Two dimensional (2D) materials with a monolayer of atoms represent an ultimate control of material dimension in the vertical direction. Molybdenum sulfide (MoS2) monolayers, with a direct bandgap of 1.8 eV, offer an unprecedented prospect of miniaturizing semiconductor science and technology down to a truly atomic scale. Recent studies have indeed demonstrated the promise of 2D MoS2 in fields including field effect transistors, low power switches, optoelectronics, and spintronics. However, device development with 2D MoS2 has been delayed by the lack of capabilities to produce large-area, uniform, and high-quality MoS2 monolayers. Here we present a self-limiting approach that can grow high quality monolayer and few-layer MoS2 films over an area of centimeters with unprecedented uniformity and controllability. This approach is compatible with the standard fabrication process in semiconductor industry. It paves the way for the development of practical devices with 2D MoS2 and opens up new avenues for fundamental research.

The transfer of synthesized 2D MoS2 films is important for fundamental and applied research. However, it is problematic to translate the well-established transfer processes for graphene to MoS2 due to different growth mechanisms and surface properties. Here we demonstrate a surface-energy-assisted process that can perfectly transfer centimeter-scale monolayer and few-layer MoS2 films from original growth substrates onto arbitrary substrates with no observable wrinkles, cracks, and polymer residues. The unique strategies used in this process include leveraging the penetration of water between hydrophobic MoS2 films and hydrophilic growth substrates to lift off the films and dry transferring the film after the lift off. This is in stark contrast with the previous transfer process for synthesized MoS2 films, which explores the etching of the growth substrate by hot base solutions to lift off the films. Our transfer process can effectively eliminate the mechanical force caused by bubble generations, the attacks from chemical etchants, and the capillary force induced when transferring the film outside solutions as in the previous transfer process, which consists of the major causes for the previous unsatisfactory transfer. Our transfer process also benefits from using polystyrene (PS), instead of poly(methyl methacrylate) (PMMA) that was widely used previously, as the carrier polymer. PS can form more intimate interaction with MoS2 films than PMMA and is important for maintaining the integrity of the film during the transfer process. This surface-energy-assisted approach can be generally applied to the transfer of other 2D materials, such as WS2.

Ridge National Laboratory, Oak Ridge, Tennessee 37831 § These authors contribute equally. AbstractSemiconductor heterostructures provide a powerful platform to engineer the dynamics of excitons for fundamental and applied interests. However, the functionality of conventional semiconductor heterostructures is often limited by inefficient charge transfer across interfaces due to the interfacial imperfection caused by lattice mismatch. Here we demonstrate that MoS 2 /WS 2 heterostructures consisting of monolayer MoS 2 and WS 2 stacked in the vertical direction can enable equally efficient interlayer exciton relaxation regardless the epitaxy and orientation of the stacking. This is manifested by a similar two orders of magnitude decrease of

4733wileyonlinelibrary.com the infl uence of substrates. [ 2 ] It has been reported that substrates may affect the luminescence effi ciency of the monolayers by inducing strain, doping, or dielectric screening. [ 2 a, e-g, 3 ] However, despite the recent progress, many important questions about the substrate effect have remained to be answered. For instance, while it is known that substrates could affect the luminescence effi ciency through multiple ways, there is no quantitative understanding for the effect of each mechanism and no knowledge on which mechanism could be dominant. More importantly, it is not clear how the effect of substrates might depend on the nature of the substrate and the physical features of the monolayers. Answers to these questions would provide useful guidance for the realization of optimal luminescence effi ciency through engineering the substrate effects. Here we quantitatively evaluate the effect of substrates on the luminescence effi ciency of monolayers MoS 2 , WS 2 , and WSe 2 and demonstrate strategies of substrate engineering to improve the effi ciency by orders of magnitude. We fi nd that the main effects of the substrate lie in doping the monolayers and facilitating defect-assisted nonradiative exciton recombinations. The doping may be from substrate-borne water moisture and the substrate itself, the former of which is much stronger than the latter for WS 2 and MoS 2 but negligible for WSe 2 . Using proper substrates can substantially mitigate the doping effect on the photoluminescence (PL), such as mica for WS 2 and MoS 2 and hexagonal boron nitride (h-BN) or polystyrene (PS) for WSe 2 . The defect-assisted recombination depends on the interaction of the defects in the monolayer such as sulfur vacancies with the substrate and may be substantially suppressed by either removing the substrate or lowering the number of defects. In this work we largely ignore the optical resonance effects associated with the substrate's geometrical features. [ 4 ] Results and DiscussionWe start with comparing the PL of suspended MoS 2 , WS 2 , and WSe 2 monolayers to those of as-grown counterparts. The monolayers were synthesized on sapphire substrates using chemical vapor deposition (CVD) processes as described previously, [ 5 ] and the suspended monolayers were prepared by manually It is demonstrated that the luminescence effi ciency of monolayers composed of MoS 2 , WS 2 , and WSe 2 is signifi cantly limited by the substrate and can be improved by orders of magnitude through substrate engineering. The substrate affects the effi ciency mainly through doping the monolayers and facilitating defect-assisted nonradiative exciton recombinations, while the other substrate effects including straining and dielectric screening play minor roles. The doping may come from the substrate and substrate-borne water moisture, the latter of which is much stronger than the former for MoS 2 and WS 2 but negligible for WSe 2 . Using proper substrates such as mica or hexagonal boron nitride can substantially mitigat...

This work reports a study on the temperature dependence of in-plane E and out-of-plane A1g Raman modes of single-layer (1L) and bi-layer (2L) MoS2 films on sapphire (epitaxial) and SiO2 (transferred) substrates as well as bulk MoS2 single crystals in a temperature range of 25-500 °C. For the films on the transferred SiO2 substrate, the in-plane E mode is only weakly affected by the substrate, whereas the out-of-plane A1g mode is strongly perturbed, showing highly nonlinear, sometimes even non-monotonic, temperature dependence on the Raman peak shift and linewidth. In contrast, for the films on the epitaxial sapphire substrate, E is affected more significantly by the substrate than A1g. This study suggests that the 2-D film-substrate coupling depends sensitively on the preparation method, and in particular on the film morphology for the transferred film. These findings are vitally important for the fundamental understanding and application of this quasi 2-D material that is expected to be supported by a substrate in most circumstances.