Answers are grounded in this paper and cite their sources.
We carried out TEM investigations on annealed Al/Ti/GaN and directly deposited TiN films on GaN. In the Al/Ti/GaN system, the annealing step gave rise to a ≈︂20 nm rough TiN film. The ohmic contact is made by TiN crystallites which are epitaxially related to GaN, the orientation relationship being (0001)GaN ∥ (111)TiN. In between, amorphous patches of 1 to 2 nm extension can be found. When the stoichiometric TiN was deposited directly on GaN, we obtained columnar TiN grains of 5 to 20 nm section which cross the whole film thickness and are rotated mostly around the [111] axis. The same epitaxial relationship is obtained and no amorphous patches are observed at the interface. When the GaN surface was locally terminated by {101-1} pyramidal features, the epitaxial relationship was slightly modified, the (0001)GaN atomic planes and {111}TiN made an angle of 5°. This is found to allow the accommodation of the local misorientation of the facets and to bring about near coincidence.
No takes yet. Share an insight, caveat, or question.
Ruterana et al. (1999) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: