Open Access
Add to BibSonomy
Abstract
Multi-wavelength fiber lasers have the advantages of low cost, easy integration with fiber systems, and wide tunable range, which is the key component of dense wavelength division multiplexing (DWDM). 2D MoTe2 micro-sheets have drawn lots of researchers’ attention due to their anisotropic layered structure and high optical adsorption. Evanescent-field (EF) devices have many applications in multi-wavelength and high-power fiber lasers. The integration of MoTe2 micro-sheets and an EF device is able to combine their advantages. Here, we report on the experimental observation of multi-wavelength solitons based on EF interaction with polarization-sensitive MoTe2 micro-sheet. Dual-wavelength conventional solitons (CSs) and switchable dual-wavelength dissipative solitons (DSs) are emitted by the proposed fiber laser. Different solitons can be effectively tuned by optimizing pump power and the state of the polarization controller. The proposed EF device and fiber laser are convenient and attractive in the fields of fiber-based sensing, frequency comb spectroscopy, and DWDM.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
There is a large variety of 2D materials, which consist of single-element layered materials (i.e., graphene, phosphorene), and mix-element layered materials (i.e., transition metal dichalcogenides, van der Waals heterostructures) [1]. The unique properties of 2D materials are high carrier mobilities, superconductivity, mechanical flexibility, good thermal conductivity, high optical and UV adsorption [2]. So 2D materials have promising applications in fields of electronics, catalysis and biosensing [3,4]. Transition metal dichalcogenides (TMDs) have important traits of singular layered structure and tunable bandgap [5–8]. MoTe2 is one of TMDs, and its crystal structure is composed of one sheet of Mo atom and two sheets of Te atom in hexagonal morphology. 2D MoTe2 film can be exfoliated from bulk MoTe2 sample with micro-mechanical cleavage method [9]. Micro-mechanical cleavage method can generate high-quality 2D sheets at ambient conditions. 2D sheets acquired by this method have the advantages of high uniformity and low pollution. So micro-mechanical cleavage method is often used on new materials that aren’t available by large-scale industrial production. High damage threshold is the inherent advantage of evanescent-field (EF) device, which can be applied in high-power system [10–12]. In order to exploit the interaction between two-dimensional (2D) materials and EF, 2D-materials are often deposited on a basal device (i.e., D-shaped fiber, tapered fiber), and then induce loss modulation in the deposited device [12,13].
Dense wavelength division multiplexing (DWDM) technology is able to transmit multiple signals with different wavelengths in the same optical fiber [14], and the number of working wavelength in DWDM has reached hundreds [15]. Since wavelength-fixed laser is difficult to meet the requirements of modern optical communication, multi-wavelength lasers have become the key component of new-generation communication networks [16], which is due to their tunable power, high stability, low cost, high flexibility and easy compatibility with optical communication systems [17]. Multi-wavelength mode-locking pulse can be achieved by many approaches [16–18]. Luo et al. had demonstrated the dual-wavelength mode-locked pulse trains with the repetition rate difference at hundreds of Hertz [16]. Yan et al. had proposed a tunable and switchable dual-wavelength Tm-doped fiber laser based on nonlinear polarization rotation (NPR) technique [18]. MoTe2 sandwiched between two fiber ferrules has been used to generate switchable and dual-wavelength conventional solitons (CSs) [19]. Numerous optical devices have been studied in the past decades, and made great contribution to the development of multi-wavelength fiber laser [20–22].
Dual-wavelength and tunable-wavelength solitons have been demonstrated in proposed fiber laser, which is based on proposed EF device with polarization-sensitive MoTe2 micro-sheet. By optimizing pump power and state of polarization controller (PC), tunable dual-wavelength dissipative solitons (DSs) with central wavelengths of 1561.4, 1565.7 nm and 1588.2, 1598.6 nm are obtained. Dual-wavelength CSs at 1532.1 and 1557.5 nm are also achieved. Experiment observations and theoretical analyses demonstrate that switchable dual-wavelength DSs are the combined result of NPR effect and MoTe2 saturable absorber (SA) effect. The proposed EF device can withstand pump power up to 550 mW and can be conveniently applied in fiber-based system. The proposed fiber laser has good mode-locking stability, and will find important applications in nonlinear optics, biomedical diagnostics systems and optical communication networks.
2. Experimental setup
The EF device used in this experiment is a D-shaped fiber (DF), which was fabricated by side-polishing single mode fiber (SMF-28e) [23]. Firstly, the SMF had been placed on a polishing wheel, and the tension imposed on SMF was approximately 0.2 N at the rotation speed of 1000 rev/min. It is paramount to keep the uniformed tension during whole polishing process, so the contact part between SMF and polishing wheel can be polished evenly [23]. The polishing process was monitored by a power meter, which was used to estimate the polishing depth. Figure 1(a) displays the microscope image of DF’s side view. It can be seen that the side width of DF is about 67.83 μm, which means that the distance from fiber core to the polishing surface is about 5.33 μm. The crystal structure of MoTe2 and image of bulk sample are shown in Fig. 1(b). The size of bulk sample is about 3 mm*5 mm, and its thickness is about 0.5 mm. The bulk sample of MoTe2 used here was prepared by chemical vapor transport method [24]. The 2D MoTe2 micro-sheets were mechanically exfoliated from bulk sample with scotch tapes [25], and deposited on polymethylmethacrylate (PMMA). Figure 1(c) depicts the image of 2D MoTe2 micro-sheet on PMMA, and it can be seen that the micro-sheet has been marked with two areas. Figure 1(d) shows the Raman spectrum of marked areas (i.e., area 1 and 2 in Fig. 1(c)), which was performed in the backscattering configuration with 633 nm excitation lasers. The characteristic phonon modes of MoTe2 in Raman spectrum are A1g at 170 cm-1, E12g at 234 cm-1 and B12g at 289 cm-1 [26,27]. The intensity ratio of these modes can be used to determine the thickness of tested MoTe2 sample [28]. When the layer number of MoTe2 sample is 1, the A1g mode has a higher Raman peak than that of E12g mode, and the B12g mode is completely disappeared [29]. Therefore, marked area 1 in Fig. 1(c) can be confirmed as monolayer MoTe2 micro-sheet. When the layer number N > 1, A1g mode and B12g mode are lower than E12g mode, and E12g mode is the highest phonon mode of MoTe2 sample. The B12g mode has maximum intensity ratio to A1g mode in bilayer sample, and then gradually decreases with respect to the increase of layer number. When the layer number of MoTe2 sample is 2, the intensity of B12g mode is higher than that of A1g mode. Then the intensity ratio between B12g mode and A1g mode became less than 1 at the layer number N > 2. When N is smaller than 10, the B12g mode and A1g mode are both visible in Raman spectrum. If N is greater than 10, the B12g mode and A1g mode are totally disappeared in the background of Raman spectrum. So, the layer number of marked area 2 in Fig. 1(c) is estimated bigger than 2, but smaller than 10 [30].
Fig. 1. (a) Microscope image of D-shaped fiber (DF)’s side view; (b) Crystal structure of MoTe2 and image of bulk MoTe2; (c) Microscope image of 2D MoTe2 films on PMMA, the sample was divided into two regions, which are marked for the test of Raman spectroscopy; (d) Raman spectrum of marked areas in Fig. 1(c) and bulk MoTe2 sample in Fig. 1(b).
The proposed EF device was fabricated with the following steps after the identification of 2D MoTe2 micro-sheet and its layer number [31,32]. Firstly, a sharp blade was mounted on a high-accuracy transfer platform, and then the blade was used to cut the PMMA/MoTe2 sample into small square (the area of square is estimated as 2500 μm2). Secondly, we located the 2D MoTe2 micro-sheet and precisely placed the sample on DF’s polishing area. Thirdly, the sample was immersed in a microscale acetone droplet, and PMMA was dissolved. Then the MoTe2 micro-sheet was floating in the microscale droplet. Since the suspended MoTe2 micro-sheet could be easily washed away by acetone, the volume of droplet must be accurately controlled. Fourthly, the MoTe2 micro-sheet was deposited on DF via optically-driven deposition method [33]. After the MoTe2 micro-sheet was separated from PMMA, we launched a continuous wave (CW) at 1550 nm with power of 10 dBm into the EF device, and monitored the status of MoTe2 deposition by checking the loss of output power. Here, the insertion loss induced by DF was 0.2 dB, and the deposition of MoTe2 increased insertion loss by 3 dB. So, the total insertion loss after deposition is 3.2 dB. Modulation depth and insertion loss increases with the growth of deposited time. In passively mode-locking lasers, modulation depth and insertion loss form a trade-off relationship [34,35]. In our experiment, we select the EF device with insertion loss of 3.2 dB, which brings the best mode-locking result. Figure 2(a) displays the schematic diagram of proposed EF device. The front and side views of proposed device indicate that the deposited area is very small compared to the size of DF, which means that the preparation of proposed device is a fine procedure. The upper image of Fig. 2(b) shows the microscope image of DF’s front view after sample deposition. It can be seen that the deposited MoTe2 micro-sheet is in the marked red circle. In order to distinguish the deposited area, we launched a 632.8 nm red-light laser into the proposed EF device, the scattering red light can be observed in Fig. 2(b)’s lower image, which means that there is a strong interaction between EF and MoTe2 micro-sheet. The upper and lower images of Fig. 2(b) are both under the magnification of 20-fold. The measurement process of nonlinear absorption is described as followed: Firstly, the linear polarization laser was divided by a 10/90 optical coupler, so the output and input could be monitored simultaneously. A PC is placed before the attenuator to adjust the polarization of input light, and the maximum and minimum non-saturable absorption reflect the nonlinear responses of TE and TM light. By gradually increasing the input power, the output loss gradually decreased, indicating that the absorption had been saturated. The experimental data is fitted with a two-level saturable absorption model α(I)= αns + α0/(1+I/Isat), where I is the intensity of input CW, α(I) is the intensity-dependent absorption, α0 is the linear limit of saturable absorption, αns is the non-saturable absorption, and Isat is the saturation intensity. Figure 2(c) shows the nonlinear saturable absorptions with respect to TE (red curve) and TM (blue curve) light, respectively. It can be seen that the corresponding modulation depth of TE light is estimated as 6.5%, and the non-saturable absorption αns of TE light is approximately 47.7%, respectively. According to the blue curve in Fig. 2(c), the corresponding modulation depth of TM light is estimated as 2.5%, and the non-saturable absorption αns of TM light is approximately 44.7%, respectively. There is no obvious polarization dependent loss (PDL) can be observed for EF device before the MoTe2 deposition, and no PDL has been reported in previous research before the material transference [36,37].
Fig. 2. (a) Schematic diagram of front and side views of proposed evanescent-field (EF) device; (b) Front image of 2D MoTe2 micro-sheet (marked area) on DF without (upper image) and with (lower image) 632.8 nm red-light laser injection; (c) Nonlinear absorption characterization of proposed EF device with TE and TM light. The square and circle are the experimental data of TE and TM light, respectively. The red and blue curves are the fit of TE and TM light, respectively.
The setup of proposed laser is shown in Fig. 3. The fiber laser includes a section of 15 m erbium-doped fiber (EDF) with group velocity dispersion (GVD) of -16 ps*nm-1*km-1 and a section of 5 m standard SMF with GVD of 17 ps*nm-1*km-1. The EDF is pumped by a 980 nm laser diode (LD) with the maximum output power of 550 mW. The tap-isolator-WDM (TIW) is an integrated optical device with combined functions of tap, isolator and WDM. The 10/90 tap port of TIW is utilized to extract the soliton from laser cavity. The isolator in the TIW is used to eliminate back-reflection of laser cavity, which usually happens at the location of 2D-material deposited area and fiber solder joints. The 980 nm pump energy is input to laser cavity by the WDM port of TIW. The proposed EF device is utilized to achieve passive mode-locking operation. In this case, the net cavity dispersion β2 is calculated as ∼0.21 ps2, and thus DS tends to be formed in the proposed fiber laser.
Fig. 3. Schematic diagram of proposed fiber laser. LD, laser diode; TIW, Integrated device with the functions of tap, isolator and WDM; EDF, erbium-doped fiber; EF device, evanescent-field device; PC, polarization controller. The inset is enlarged image of proposed EF device.
3. Experimental results
CW was observed at the pump power of 20 mW. Then we gradually increased the pump power to 30 mW, a self-starting mode-locking operation with single wavelength DS was obtained [38]. By properly adjusting the state of PC and pump power, two tunable dual-wavelength DSs at pump power of 50 mW are depicted in Fig. 4(a). The DSs have quasi-rectangular profiles of spectrum, which is the representative feature of DS. The central wavelength and quasi-rectangular bandwidth of four DSs are measured as 1561.4 and 8.8 nm, 1565.7 and 7.5 nm, 1588.2 and 15.4 nm, 1598.6 and 19.4 nm, respectively. It can be seen that two dual-wavelength DSs have smaller bandwidth at shorter wavelength than that of DSs at longer wavelength, which are due to the NPR-induced spectral filter and gain profile of employed EDF. The corresponding radio-frequency (RF) spectrum of dual-wavelength operation at 1561.4 and 1588.2 nm (i.e., blue curve in Fig. 4(a)) is shown in Fig. 4(b). The fundamental repetition rates of both DSs are about 10.27251 MHz with a peak-to-background ratio of ∼57 dB for longer wavelength DS and 10.27297 MHz with a peak-to-background ratio of ∼52 dB for shorter wavelength DS. Figure 4(c) shows the corresponding oscilloscope trace of dual-wavelength operation at 1561.4 and 1588.2 nm. The pulse-to-pulse separation is about 97 ns, which is equal to the cavity length of 20 m. When two solitons propagate in the same laser cavity with different wavelength, the repetition rate difference between two solitons is expressed as Δƒ=Δt/t2=LDΔλ1/t2, where Δƒ is the difference of repetition rate, Δλ1 is the difference of wavelength, t is the round-trip time, D is the average dispersion of employed fiber. Here, D can be calculated as -7.8 ps/(nm·km), t is about 97 ns, L and Δλ1 is ∼20 m and 26.8 nm, respectively. Then, Δf is calculated as -441 Hz. In order to reduce the duplicate information, only the corresponding RF spectrum and oscilloscope trace of dual-wavelength operation at 1561.4 and 1588.2 nm are given in Fig. 4. The corresponding RF spectrum and oscilloscope trace of dual- wavelength operation at 1565.7 and 1598.6 nm have slightly difference to that of Figs. 4(b) and 4(c).
Fig. 4. Experimental results of tunable dual-wavelength dissipative solitons (DSs). (a) Optical spectrum of DSs; (b) Radio-frequency (RF) spectrum of the blue curve in Fig. 4(a); (c) Oscilloscope trace of the blue curve in Fig. 4(a).
It is worth noting that the other operations also can be obtained in the proposed fiber laser by altering the net dispersion of resonant cavity. Here, the net dispersion had become -0.13 ps2 by adding 15 m SMF to the original laser cavity. Figure 5 shows the optical spectrum, RF spectrum and oscilloscope trace of dual-wavelength CSs, as well as the optical spectrum of tunable single-wavelength CSs. Figure 5(a) depicts the dual-wavelength CSs (i.e., red curve) with central wavelengths of 1532.1 and 1557.5 nm, tunable CSs (i.e., green curve and blue curve) with central wavelength of 1543.1 and 1563.9 nm. Sidebands are the typical feature of CS spectrum [39]. The first peak on the left side of green curve in Fig. 5(a) is a Kelly sideband, and the second peak on its left side is parasitic CW. Figure 5(b) shows the corresponding RF spectrum of dual-wavelength CSs. The fundamental repetition rates of 1532.1 nm and 1557.5 nm soliton are about 5.87280 MHz with a peak-to-background ratio of ∼50 dB and 5.87288 MHz with a peak-to-background ratio of ∼56 dB, respectively. Figure 5(c) shows the corresponding oscilloscope trace of dual-wavelength CSs. The pulse-to-pulse separation is about 170 ns, which is equal to the length of new cavity (i.e., 35 m). Here, D can be calculated as -2.9 ps/(nm·km), t is about 170 ns, L and Δλ1 is ∼35 m and 25.4 nm, respectively. Then, Δf is calculated as 88 Hz, which is coincided with experimental result in Fig. 5(b).
Fig. 5. Experimental results of dual-wavelength and tunable single-wavelength conventional solitons (CS). (a) Optical spectrum of CSs; (b) RF spectrum of red curve in Fig. 5(a); (c) Oscilloscope trace of red curve in Fig. 5(a).
Figure 6 displays the typical experimental results of single-wavelength DS and CS. It can be seen the optical spectrum and autocorrelation (AC) trace of DS in Figs. 6(a) and 6(c). The optical spectrum and AC trace of single-wavelength CS are shown in Figs. 6(b) and 6(d), respectively. According to Figs. 6(a) and 6(c), the spectrum bandwidth and full width at half maximum (FWHM) of AC trace are measured to be 11 nm and 15.4 ps (Gauss fit), respectively. The time bandwidth product (TBP) of DS is calculated as 14.9, which indicates that the DS is highly chirped. On the other hand, the spectrum bandwidth and FWHM of AC trace in Figs. 6(b) and 6(d) are measured to be 4.2 nm and 1.9 ps (Sech2 fit), respectively. The TBP of CS is calculated as 0.63, which is much smaller than that of DS. The pulse can be compressed by transmitting in a section of external SMF. According to the following equation, $TBP = c\varDelta {\lambda _2}\varDelta \tau /\lambda _s^2$, where c represents the light speed in vacuum, λs represents the central wavelength of pulse, Δλ2 represents the spectrum bandwidth of soliton, and Δτ represents the pulse duration. Since the transform limited of Gauss fitting is 0.44, the minimum duration of DS can be calculated as 0.34 ps. The minimum duration of CS can be calculated as 0.60 ps, which is larger than that of DS.
Fig. 6. Experimental results of typical single-wavelength DS and CS. (a) Optical spectrum of DS; (b) Optical spectrum of CS; (c) Autocorrelation (AC) trace of DS; (d) AC trace of CS.
4. Discussion
These experiment results demonstrate the intense interaction between light field and 2D MoTe2 micro-sheet, which results the generation of multiple dual-wavelength solitons. According to the defect state theory of 2D materials, the appearance of defect state in 2D material film is inevitable, and the defect state will renew the electronic and optical properties of 2D material film [40,41]. Here, the defect state in 2D MoTe2 micro-sheet leads to its polarization- sensitive property. The same property had also been observed in other 2D material film, which had been applied in polarizer or modulator [42,37]. The polarization-sensitive device and PC have formed a NPR-induced spectral filter in proposed fiber laser [43–45]. Therefore, NPR effect and saturable absorption effect work as two mode-lockers, and realize multiple mode-locking operations in the proposed fiber laser.
The properties of NPR-induced filter can be tuned by optimizing PC’s polarization state and pump power. The transmission of NPR-induced filter can be calculated by the following equation [38]:
Fig. 7. Transmission with respect to instantaneous power of soliton, linear phase delay and wavelength. (a) Transmission with respect to instantaneous power of soliton, red and green curves are initial phase delay at 0.1 π and 0.3 π when 0◦<α<45◦, blue and purple dash curves are initial phase delay at 0.1 π and 0.3 π when 45◦<α<90◦; (b) Transmission with respect to linear phase delay, red and green curves are instantaneous power at 10 W and 15 W when 0◦<α<45◦, blue and purple dash curves are instantaneous power at 10 W and 15 W when 45◦<α<90◦; (c) Transmission with respect to wavelength, red and green curves are instantaneous power at 5 W and 10 W when 0◦<α<45◦, blue and purple dash curves are instantaneous power at 5 W and 10 W when 45◦<α<90◦; (d) Transmission with respect to wavelength, red and green curves are initial phase delay at 0.1 π and 0.3 π when 0◦<α<45◦, blue and purple dash curves are initial phase delay at 0.1 π and 0.3 π when 45◦<α<90.
The difference in PC’s polarization state and pump power will generate pulses with different bandwidths and central wavelengths. When the proposed fiber laser worked at different PC states and pump power, the background noise experienced different loss. Then the difference in cavity loss led to different intensity, fiber nonlinearity and saturable absorption effect, which resulted multiple dual-wavelength solitons. So, the feature of soliton is decided by the combined effect of SA and NPR. On the other hand, when we used the undeposited EF device in proposed laser, there is none soliton, which indicates that the MoTe2 micro-sheet plays the key role as mode-locker. Meanwhile, the proposed EF device also had been replaced by other mode-lockers (i.e., MoTe2 micro-sheet sandwiched between two fiber ferrules or CNT deposited on DF). When the proposed EF device had been replaced by MoTe2 micro-sheet sandwiched between two fiber ferrules, wavelength-tunable operation can be obtained but without dual-wavelength operation, which means that dual-wavelength operation in the proposed laser is attributed to the interaction between EF and MoTe2 micro-sheet. When the proposed EF device had been replaced by CNT deposited on DF, dual-wavelength operation can be obtained but without wavelength-tunable operation. Unlike the MoTe2 based SA, CNTs based SA is a polarization-independent mode-locker [46], so the laser mode-locked by CNTs based SA is independent to polarization perturbation, which means that the generation of soliton is under the effect of single mode-locker. The above experiment results confirm that wavelength-tunable operation is attributed to the polarization-sensitive property of MoTe2 micro-sheet. There are some parasitic-CW sidebands and high-peak Kelly sidebands in Fig. 5(a) and Fig. 6(b), which indicates that the soliton become unstable and its pulse duration is near the minimum possible value [39,47,48]. In order to prevent the instability of sidebands, reducing pump power is the most effective method [49,50].
5. Conclusion
The proposed device based on EF and 2D-material film has been applied in fiber laser, and tunable dual-wavelength DSs and CSs have been generated by proposed laser. Multiple dual-wavelength solitons can be obtained by exploiting the combined effect of SA and NPR. The properties of NPR-induced filter can be tuned by changing PC’s polarization state and pump power. The switchable dual-wavelength DSs have central wavelengths of 1561.4, 1588.2 nm and 1565.7, 1598.6 nm with corresponding spectral bandwidths of 8.8, 15.4 nm and 7.5, 19.4 nm, respectively. The dual-wavelength CSs have central wavelengths of 1532.1 and 1557.5 nm with corresponding spectral bandwidths of 1.7 and 1.8 nm, respectively. The preparation of proposed EF device is simple, easy-adjustable and low cost, which may benefit a wide range of novel 2D materials. The proposed all-fiber laser can simultaneously provide tunable dual-wavelength DSs and CSs, which will greatly promote the development of DWDM networks.
Funding
National Natural Science Foundation of China (61805212, 51802245, 61405151, 11604252); National Key Research and Development Program of China (2017YFF0204901); Natural Science Foundation of Shaanxi (20180418); Specific scientific research project in Shaanxi Province department of Education (20JK0662); National Key Research and Development Program of China (2018YFB2200500); Natural Science Foundation of Zhejiang Province (Y19F050039); Key Laboratory of Artificial Microstructure, Ministry of Education, Wuhan University (13022019af002) .
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. E. P. Nguyen, C. Silva, and A. Merkoi, “Recent advancement in biomedical applications on the surface of two-dimensional materials: from biosensing to tissue engineering,” Nanoscale 12(37), 19043–19067 (2020). [CrossRef]
2. S. Yan, X. Zhu, J. Dong, and S. Xiao, “2D materials integrated with metallic nanostructures: Fundamentals and optoelectronic applications,” Nanophotonics 9(7), 0074 (2020). [CrossRef]
3. P. Bollella, G. Fusco, C. Tortolini, S. Gabriella, and R. Antiochia, “Beyond graphene: electrochemical sensors and biosensors for biomarkers detection,” Biosensors and Bioelectronics 89, 152–166 (2017). [CrossRef]
4. G. Imrich, J. Pekárek, K. Jaroslav, S. Vojtech, S. Ali, N. Pavel, and P. Martin, “Monoelemental 2D materials-based field effect transistors for sensing and biosensing: phosphorene, antimonene, arsenene, silicene, and germanene go beyond graphene,” Trend Anal. Chem. 105, 251–262 (2018). [CrossRef]
5. Y. Yue, Y. Feng, J. Chen, D. Zhang, and F. Wei, “Two-dimensional large-scale bandgap-tunable monolayer MoS2(1−x)Se2x/graphene heterostructures for phototransistors,” J. Mater. Chem. C 5(24), 5887–5896 (2017). [CrossRef]
6. S. Susarla, A. Kutana, J. A. Hachtel, V. Kochat, A. Apte, R. Vajtai, J. Idrobo, B. I. Yakobson, C. S. Tiwary, and P. M. Ajayan, “2D materials: quaternary 2D transition metal dichalcogenides (TMDs) with tunable bandgap,” Adv. Mater. 29(35), 70255 (2017). [CrossRef]
7. D. H. Keum, S. Cho, J. H. Kim, D. Choe, and Y. Lee, “Bandgap opening in few-layered monoclinic MoTe2,” Nat. Phys. 11(6), 482–486 (2015). [CrossRef]
8. T. Nawz, A. Safdar, M. Hussain, D. S. Lee, and M. Siyar, “Graphene to advanced MoS2: a review of structure, synthesis, and optoelectronic device application,” Crystals 10(10), 902 (2020). [CrossRef]
9. L. Torres, L. Gomez, and A. Carlos, “Optimization of micromechanical cleavage technique of natural graphite by chemical treatment,” Graphene 03(01), 1–5 (2014). [CrossRef]
10. T. Thio, “A bright future for subwavelength light sources,” Am. Sci. 94(1), 40–47 (2006). [CrossRef]
11. P. L. Marston, “Scattering of acoustic evanescent waves by circular cylinders: partial wave series solution,” J. Acoust. Soc. Am. 111(5), 2378 (2002). [CrossRef]
12. D. Mao, Y. Wang, C. Ma, L. Han, B. Jiang, X. Gan, S. Hua, W. Zhang, T. Mei, and J. Zhao, “WS2 mode-locked ultrafast fiber laser,” Sci. Rep. 5(1), 7965 (2015). [CrossRef]
13. A. Schell, H. Takashima, T. Tran, I. Aharonovich, and S. Takeuchi, “Coupling quantum emitters in 2D materials with tapered fibers,” ACS Photonics 4(4), 761–767 (2017). [CrossRef]
14. T. Verolet, G. Aubin, Y. Lin, C. Browning, and A. Ramdane, “Mode locked laser phase noise reduction under optical feedback for coherent DWDM communication,” J. Lightwave Technol. 38(20), 5708–5715 (2020). [CrossRef]
15. R. Hashemi, H. Beyranvand, and H. Rabbani, “Joint channel power and amplifier gain optimization in coherent DWDM systems,” Optics Communications 475, 126212 (2020). [CrossRef]
16. T. H. Luo, T. S. Tuan, H. P. T. Saini, Y. Nguyen, and Ohishi, “Tunable and switchable all-fiber dual-wavelength mode locked laser based on lyot filtering effect,” Opt. Express 27(10), 14635 (2019). [CrossRef]
17. Z. Yan, X. Li, Y. Tang, P. Shum, X. Yu, Y. Zhang, and Q. Wang, “Tunable and switchable dual-wavelength Tm-doped mode-locked fiber laser by nonlinear polarization evolution,” Opt. Express 23(4), 4369 (2015). [CrossRef]
18. S. Pan and C. Lou, “Stable multiwavelength dispersion-tuned actively mode-locked erbium-doped fiber ring laser using nonlinear polarization rotation,” IEEE Photon. Tech. Lett. 18(13), 1451–1453 (2006). [CrossRef]
19. A. Yasim, “Switchable and dual-wavelength ultrafast fibre lasers with an MoTe2-based saturable absorber,” J. Mod. Opt. 67(4), 367–373 (2020). [CrossRef]
20. Y. Li, Y. Cao, L. Gao, L. Huang, and T. Zhu, “Fast spectral characterization of optical passive devices based on dissipative soliton fiber laser assisted dispersive fourier transform,” Phys. Rev. Appl. 14(2), 024074 (2020). [CrossRef]
21. J. Wang and J. Dong, “Optical waveguides and integrated optical devices for medical diagnosis, health monitoring and light therapies,” Sensors 20(14), 3981 (2020). [CrossRef]
22. L. Wu, Y. Dong, J. Zhao, D. Ma, W. Huang, Y. Zhang, Y. Wang, X. Jiang, Y. Xiang, J. Li, Y. Feng, J. Xu, and H. Zhang, “Kerr nonlinearity in 2D graphdiyne for passive photonic diodes,” Adv. Mater. 31(14), 1807981 (2019). [CrossRef]
23. F. DeJun, Z. Mao, G. Liu, L. Lu, and J. Dong, “D-shaped plastic optical fiber sensor for testing refractive index,” IEEE Sens. J. 14(5), 1673–1676 (2014). [CrossRef]
24. D. Hu, G. Xu, L. Xing, X. Yan, J. Wang, J. Zheng, Z. Lu, P. Wang, X. Pan, and L. Jiao, “Two-dimensional semiconductors grown by chemical vapor transport,” Angew. Chem. 129(13), 3665–3669 (2017). [CrossRef]
25. P. Budania, P. T. Baine, J. H. Montgomery, D. W. McNeill, S. Mitchell, M. Modreanu, and P. Hurley, “Comparison between scotch tape and gel-assisted mechanical exfoliation techniques for preparation of 2D transition metal dichalcogenide flakes,” Micro Nano Lett. 12(12), 970–973 (2017). [CrossRef]
26. H. Zhang, W. Zhou, X. Li, J. Xu, Y. Shi, B. Wang, and F. Miao, “High temperature raman investigation of few-layer MoTe2,” Appl. Phys. Lett. 108(9), 147–5983 (2016). [CrossRef]
27. N. R. Pradhan, D. Rhodes, S. Feng, X. Yan, and L. Balicas, “Field-effect transistors based on few-layered alpha-MoTe2,” Physics 8(6), 5911–5920 (2014). [CrossRef]
28. M. Yamamoto, S. Wang, M. Ni, Y. Lin, S. Li, S. Aikawa, W. Jian, K. Ueno, K. Yashi, and K. Tsukagoshi, “Strong enhancement of Raman scattering from a bulk-inactive vibrational mode in few-layer MoTe2,” ACS Nano 8(4), 3895–3903 (2014). [CrossRef]
29. X. Luo, Y. Zhao, J. Zhang, M. Toh, C. Kloc, Q. Xiong, and S. Quek, “Effects of lower symmetry and dimensionality on Raman spectra in two-dimensional WSe2,” Phys. Rev. B 88(19), 195313 (2013). [CrossRef]
30. C. Ruppert, O. Aslan, and T. F. Heinz, “Optical properties and band gap of single- and few-layer MoTe2 crystals,” Nano Lett. 14(11), 6231–6236 (2014). [CrossRef]
31. W. Li, B. Chen, C. Meng, W. Fang, Y. Xiao, X. Li, Z. Hu, Y. Xu, L. Tong, and H. Wang, “Ultrafast all-optical graphene modulator,” Nano Lett. 14(2), 955–959 (2014). [CrossRef]
32. B. Chen, C. Meng, Z. Yang, W. Li, and L. Tong, “Graphene coated ZnO nanowire optical waveguides,” Opt. Express 22(20), 24276–24285 (2014). [CrossRef]
33. K. Kashiwagi and S. Yamashita, “Deposition of carbon nanotubes around microfiber via evanescent light,” Opt. Express 17(20), 18364 (2009). [CrossRef]
34. J. Jeon, J. Lee, and H. L. Ju, “Numerical study on the minimum modulation depth of a saturable absorber for stable fiber laser mode locking,” J. Opt. Soc. Am. B 32(1), 31–37 (2015). [CrossRef]
35. W. J. Liu, M. Liu, Y. Ouyang, H. Hou, and Z. Wei, “CVD-grown MoSe2 with high modulation depth for ultrafast mode-locked erbium-doped fiber laser,” Nanotechnology 29(39), 394002 (2018). [CrossRef]
36. J. D. Zapata, D. Steinberg, L. Saito, R. de Oliveira, A. Cárdenas, and E. A. Thoroh de Souza, “Efficient graphene saturable absorbers on D-shaped optical fiber for ultrashort pulse generation,” Sci. Rep. 6(1), 20644 (2016). [CrossRef]
37. R. L. de Oliveira and C. J. de Matos, “Graphene based waveguide polarizers: in-depth physical analysis and relevant parameters,” Sci. Rep. 5(1), 16949 (2015). [CrossRef]
38. A. Komarov, H. Leblond, and F. Sanchez, “Multistability and hysteresis phenomena in passively mode-locked fiber lasers,” Phys. Rev. A 71(5), 053809 (2005). [CrossRef]
39. S. Kelly, “Characteristic sideband instability of periodically amplified average soliton,” Electron. Lett. 28(8), 806–807 (1992). [CrossRef]
40. F. Banhart, J. Kotakoski, and A. V. Krasheninnikov, “Structural defects in graphene,” ACS Nano 5(1), 26–41 (2011). [CrossRef]
41. J. Hong, Z. Hu, M. Probert, K. Li, D. Lv, X. Yang, L. Gu, N. Mao, Q. Feng, and L. Xie, “Exploring atomic defects in molybdenum disulphide monolayers,” Nat. Commun. 6(1), 6293 (2015). [CrossRef]
42. E. J. Lee, S. Choi, H. Jeong, N. Park, W. Yim, M. Kim, J. Park, S. Son, S. Bae, and S. Kim, “Active control of all-fibre graphene devices with electrical gating,” Nat. Commun. 6(1), 6851 (2015). [CrossRef]
43. W. S. Man, H. Y. Tam, M. S. Demokan, P. K. A. Wai, and D. Y. Tang, “Mechanism of intrinsic wavelength tuning and sideband asymmetry in a passively mode-locked soliton fiber ring laser,” J. Opt. Soc. Am. B 17(1), 28–33 (2000). [CrossRef]
44. D. Yan, X. Li, S. Zhang, M. Han, H. Han, and Z. Yang, “L-band wavelength-tunable dissipative soliton fiber laser,” Opt. Express 24(2), 739 (2016). [CrossRef]
45. K. Ozgören and F. Ö. Ilday, “All-fiber all-normal dispersion laser with a fiber-based lyot filter,” Opt. Lett. 35(8), 1296–1298 (2010). [CrossRef]
46. A. G. Rozhin, Y. Sakakibara, and S. Namiki, “Sub-200-fs pulsed erbium-doped fiber laser using a carbon nanotube-polyvinyl alcohol mode locker,” Appl. Phys. Lett. 88(5), 051118 (2006). [CrossRef]
47. N. Smith, K. Blow, and I. Andonovic, “Sideband generation through perturbations to the average soliton model,” J. Lightwave Technol. 10(10), 1329–1333 (1992). [CrossRef]
48. J. Gordon, “Dispersive perturbations of solitons of the nonlinear Schroedinger equation,” J. Opt. Soc. Am. B 9(1), 91–97 (1992). [CrossRef]
49. M. Dennis and I. Duling III, “Experimental study of sideband generation in femtosecond fiber lasers,” IEEE J. Quantum Electron. 30(6), 1469–1477 (1994). [CrossRef]
50. D. Jones, Y. Chen, H. Haus, and E. Ippen, “Resonant sideband generation in stretched-pulse fiber lasers,” Opt. Lett. 23(19), 1535–1537 (1998). [CrossRef]
