Materials synthesis: Two-dimensional gallium nitride.

The groundbreaking discovery1 that graphite could be cleaved into its 2D building block, graphene, has led to similar efforts directed towards other layered 3D solids. In fact, a wide variety of 2D materials have been discovered over the past decade: phosphorene, transition metal dichalcogenides and MXenes (transition metal carbides, nitrides or carbonitrides) to name just a few2,3. Most such 2D materials are derived from 3D van der Waals solids (layered materials comprising 2D sheets stacked on each other and held together by weak van der Waals forces)3. Such van der Waals bonded layered compounds can be exfoliated by mechanical, chemical or electrochemical methods to isolate their 2D form. But there is no clear path so far to extract 2D forms of 3D crystals that lack this layered structure. For example, tetrahedrally coordinated bulk crystals such as wurtzite gallium nitride (GaN) cannot be easily cleaved, owing to unsaturated dangling bonds on the surface. If one could passivate such dangling bonds, then synthesis of 2D forms of GaN could in fact be possible. Writing in Nature Materials, Al Balushi and co-workers4 report that they have done just this through a migration-enhanced encapsulation growth (MEEG) technique that uses graphene as a capping or stabilizing layer during synthesis. The unique aspect of MEEG is that the GaN growth occurs in between the growth substrate (silicon carbide, SiC) and a graphene capping bilayer (Fig. 1a–c). This graphene capping sheet is formed by sublimation of Si from the SiC substrate followed by hydrogenation. Trimethylgallium is decomposed on the graphene/SiC surface, causing gallium atoms to penetrate through the graphene into the interstitial space between the SiC substrate and the graphene bilayer as illustrated schematically in Fig. 1a. Al Balushi and co-workers4 show that defects (such as carbon vacancies), wrinkles (tears) and large gallium metal islands act as the favoured sites for gallium to penetrate through the graphene. Once gallium atoms intercalate into the graphene/SiC interface, they appear to be stabilized as a bilayer of gallium, probably because of the highly passivated nature of the hydrogenated SiC/graphene interface. In the final growth steps, atomic nitrogen (created from decomposition of ammonia) also penetrates through the defective graphene capping sheet (Fig. 1b) and reacts with the intercalated gallium to form a patchwork of 2D GaN islands (Fig. 1c,d). The presence of the graphene bilayer is indispensable for successful GaN growth in its 2D form; in fact, the researchers showed that GaN grown on the SiC substrate without the graphene capping sheet resulted only in 3D islands. Aberration-corrected scanning transmission electron microscopy indicates that the 2D GaN is in a buckled state owing to the passivation of surface dangling bonds. Further, the stoichiometry of 2D GaN is MATERIALS SYNTHESIS