The layered transition metal dichalcogenides (TMDs) MX 2 (M = Mo, W; X = S, Se, Te), a class of graphene-like two-dimensional materials, have attracted significant interest because they demonstrate quantum confinement at the single-layer limit 13 . As with graphene, these layered materials can be easily exfoliated mechanically to provide monolayers 3-7,14-16 and assume a hexagonal honeycomb structure in which the M and X atoms are located at alternating corners of the hexagons. However, unlike graphene, which has a gapless Dirac cone band structure, MX 2 has a rather large bandgap, making these materials more versatile as candidates for thin, flexible device applications and useful for a variety of other applications including lubrication 16 , catalysis 17 , transistors 18 and lithium-ion batteries 19 . Most interestingly, an indirect to direct bandgap transition in the monolayer limit has been predicted theoretically and supported experimentally by optical measurements [3][4][5]9,12 . Because of the direct bandgap, monolayer MX 2 is favourable for optoelectronic applications5 and field-effect transistors 15,16,18 . Furthermore, both the conduction and valence bands have two energy degenerate valleys at corners of the first Brillouin zone, making it viable to optically control the charge carriers in these valleys and suggesting the possibility of valley-based electronic and optoelectronic applications 3,6-8 .Despite these exciting developments, direct experimental verification of the novel band structure at the monolayer limit remains lacking. Furthermore, for many applications, it is vital to manufacture high-quality epitaxial films with controllable methods such as chemical vapour deposition (CVD) or molecular beam epitaxy (MBE) 20,21 .
