Two-dimensional (2D) semiconductors such as single- and few-layer molybdenum disulphide(MoS2) are promising building blocks for prospect flexible, transparent and low power electronics. Due to an electronic bandgap of the order of ~1.8 eV and atomic-scale tenuity of 6.5 Å, field-effect transistors (FETs) based on single-layer MoS2 exhibit high current Ion/Ioff-ratios and low power dissipation. Encapsulating single-layerMoS2 within a dielectric layer protects it from exposure to ambient air and improves device performance. By investigating the performance limits of such devices we demonstrate that they can reproducibly sustain several 100 µA/µm, saturate at high bias and exhibit break down current densities 50× larger than Cu. In addition, we investigate high-frequency operation of high-performance MoS2 FETs based on one- to few-layers. They exhibit current-, power- and voltage-gain in the GHz-range. Improving our understanding of contacts formed between metals and 2D semiconductors is important for the advancement of this technology because Schottky barriers impede device performances. The common assumption when modeling these contacts is that the thermionic emission model for a three-dimensional system is valid also in 2D. Here, we show that the understanding of 2D semiconductor/metal junctions can be improved by considering imageforce barrier-lowering and neglecting band bending at the interface. With this modifications, we can reproduce the output characteristics of 2D MoS2 FETs with excellent accuracy. The model is applicable over the whole range of investigated parameters such as temperature, doping and applied bias. While single-layer MoS2 attracted interest due to its broken inversion symmetry and spin/valley coupling, few-layer MoS2 is considered a more viable option for technological applications where its higher mobility and lower contact resistance offer an advantage. However, so far it remains unclear whether few-layers are intrinsically superior or if the difference is due to environmental effects. Here, we provide a systematic comparison of the field-effect mobility in single-, double- and trilayer MoS2 after thorough in-situ annealing. It shows that a single-layer exhibits the highest mobilities, while it is most sensitive to ambient air. In conclusion, this thesis offers important contributions towards the understanding of the performance limits, Schottky barrier height and field-effect mobility in encapsulated and bare 2D MoS2. The presented results support the feasibility of novel flexible, transparent and low power electronics based on this material.