The 2D carbon material graphene, which has unusual electronic and optical properties, was first created via exfoliation. Initially, single layers of graphite were extracted and subsequently peeled off using Scotch tape. It is, however, essential for 2D layered materials to be thermodynamically stable in air, which graphene is not. 2D materials that can provide this stability are highly sought after. Due to their semiconducting and photoelectric nature, layers of molybdenum disulfide (MoS2), one of a class of transition metal dichalcogenides (TMDCs), are beginning to attract the attention of scientists and engineers. One layer of MoS2 forms a stack comprising a layer of molybdenum atoms between two layers of sulfur atoms.
By decreasing its thickness, the bandgap of MoS2 can be tuned (from a direct to an indirect semiconductor) with much greater ease than graphene. This energy level tunability allows a rich variety of spectral operation for application in photoelectronics. Moreover, the stronger light-matter interaction and higher electronic mobility of MoS2 make it a possible replacement for graphene in optoelectronic device manufacturing.
A saturable absorber, in which the saturable absorption (SA) effect is exploited, represents one important application for MoS2. Under this effect, the absorption of light decreases with increasing light intensity, representing a crucial component for achieving pulse operation in a laser. Photon absorption causes some of the electrons orbiting the molecules to be kicked into higher energy states, and while the electrons remain in these states (a duration of ∼1fs in MoS2), photons can pass through. In 2013, Wang and co-workers found that MoS2 dispersions exhibit a much stronger SA response than has been observed in graphene dispersions.1
Optical deposition enables the coupling of a 2D-layered-structure material into a fiber laser.2 Figure 1 shows a schematic diagram of our experimental setup. When the light provided by a laser diode is pumped through an optical fiber tail, photons become immersed in the material suspension. Under laser illumination, thermal convection flow and thermal diffusion cause MoS2 flakes to move along the temperature gradient and toward the fiber surface, resulting in the material becoming optically trapped.2 This fiber is then connected to another clear fiber tail through a flange, producing a fiber-pigtailed saturable absorber (see Figure 2).3
Figure 1. Experimental setup for photodeposition. Laser light (975nm) is shone onto the molybdenum disulfide (MoS2) dispersion through a fiber tail. LD: Laser diode.
Figure 2. Schematic of the ytterbium-doped fiber (YDF) laser, passively mode-locked by the (MoS2) absorber. WDM: Wavelength division multiplexer. PC: Polarization controller. PI-ISO: Polarization-independent isolator. SMF: Single-mode fiber.
We recently succeeded in coupling few-layer MoS2 onto a fiber tail surface using optical deposition, enabling us to obtain a mode-locked pulse output from a fiber laser at a wavelength of 1054nm. Figures 2 and 3 show a schematic of the fiber laser, mode-locked by the MoS2 saturable absorber, and properties of the pulse, respectively.
Figure 3. The laser mode-locking performance by the MoS2saturable absorber: (a) the SA response of the fiber-pigtailed MoS2saturable absorber; (b) and (c) the oscilloscope tracings; (d) the optical spectrum; (e) the radio frequency (RF) spectral profile; and (f) the wideband RF spectrum. a.u.: Arbitrary units.
Although the photon energy at 1054nm is not enough to initiate photon transition within the materials, few-layer MoS2 does possess wavelength-insensitive SA responses, due to its special molecular structure. The exfoliated MoS2 nanoplatelet sample comprises a mixture of 1T (metallic) and 2H (semiconducting) phases. The 1T phase usually predominates in as-exfoliated samples due to doping by impurities, which gives rise to similar broadband performance. If the MoS2 can be rendered predominantly 2H, its absorption at resonance energies will be stronger.3
The stability and robustness issues of TMDCs, however, present a significant problem under exposure to high-power laser illumination. Polymethyl methacrylate (PMMA) is indispensable for protecting few-layer MoS2 from vertical transmission under strong optical power density. In our experiment, the composite (MoS2/PMMA) adheres to a fiber tail with a mode-field diameter of several micrometers. In principle, pure MoS2 cannot tolerate laser illumination higher than 100mW. By implementing PMMA, the composite material can withstand illumination of up to 500mW. This still, however, represents a serious limitation regarding its potential application in practical optical devices. A taper fiber with an evanescent field may represent an effective solution to this problem. Additionally, few-layer MoS2 adhered to the waist of the taper fiber may enable damage by strong optical power densities to be avoided. Such an optical device—see Figure 4—could bear 1W laser injection without damage.4
Figure 4. Schematic diagram of the taper fiber and the YDF laser passively mode-locked by the MoS2-taper-fiber saturable absorber.
In summary, we have proposed a convenient and practical way to overcome the very low optical damage threshold of 2D semiconducting TMDCs, simply by adopting a lateral interaction scheme. We anticipate that, stimulated by this technological innovation, researchers may propose new types of light interaction modes with 2D materials, particularly regarding their integration into various structures (e.g., silicon waveguides). In addition to solving the problems concerning optical damage, further research could also lead to new physics describing the way in which light propagates along and interacts with a 2D semiconducting surface. We believe that this may eventually transform our perspectives on 2D optoelectronics and open up a new test bed with unprecedented opportunities for novel optoelectronic devices.
In future work, we aim to optimize the MoS2 SA modulation depth, which could effectively suppress the wave-break effect in a mode-locked fiber laser, showing promise for next-generation lasers. Using a microfiber-based MoS2saturable absorber, we have experimentally demonstrated both high optical nonlinearity and SA. A device such as this could have wide application in fields such as ultrafast nonlinear optics.
Nanyang Technological University (NTU)
Feng Luan obtained his BSc in physics from Peking University, China, in 1999. He subsequently joined the photonics material group at the University of Bath, England, where he received his PhD in 2005. He is currently an assistant professor in the school of electrical and electronic engineering at NTU.
Han Zhang graduated from Wuhan University, China, with a BSc in physics in 2006, and from Nanyang Technological University, Singapore, with a PhD in electrical and electronic engineering in 2010. He is now a professor of photonics at Shenzhen University. His research includes photonics based on 2D materials, ultrafast lasers, and nonlinear optics. His publications have been cited more than 3600 times, and his H-factor is 30.
1. K. Wang, J. Wang, J. Fan, M. Lotya, A. O'Neill, D. Fox, Y. Feng, Ultrafast saturable absorption of 2D MoS2 nanosheets, ACS Nano 7(10), p. 9260-9267, 2013.
2. H. Kim, J. Cho, S. Y. Jang, Y. W. Song, Deformation-immunized optical deposition of graphene for ultrafast pulsed lasers, Appl. Phys. Lett. 98(2), p. 021104, 2011.
3. H. Zhang, S. B. Lu, J. Zheng, J. Du, S. C. Wen, D. Y. Tang, K. P. Loh, Molybdenum disulfide (MoS2) as a broadband saturable absorber for ultra-fast photonics, Opt. Express 22(6), p. 7249-7260, 2014.
4. J. Du, Q.-K. Wang, G.-B. Jiang, C.-W. Xu, C.-J. Zhao, Y.-J. Xiang, Y. Chen, Ytterbium-doped fiber laser passively mode locked by few-layer molybdenum disulfide (MoS2
) saturable absorber functioned with evanescent field interaction, Sci. Rep.
4, p. 6346, 2014. doi:10.1038/srep06346