Search for: All records

The simplest picture of excitons in materials with atomic-like localization of electrons is that of Frenkel excitons, where electrons and holes stay close together, which is associated with a large binding energy. Here, using the example of the layered oxide V₂O₅, we show how localized charge-transfer excitations combine to form excitons that also have a huge binding energy but, at the same time, a large electron-hole distance, and we explain this seemingly contradictory finding. The anisotropy of the exciton delocalization is determined by the local anisotropy of the structure, whereas the exciton extends orthogonally to the chains formed by the crystal structure. Moreover, we show that the bright exciton goes together with a dark exciton of even larger binding energy and more pronounced anisotropy. These findings are obtained by combining first principles many-body perturbation theory calculations, ellipsometry experiments, and tight binding modelling, leading to very good agreement and a consistent picture. Our explanation is general and can be extended to other materials.

 

A predicted type-II staggered band alignment with an approximately 1.4 eV valence band offset at the ZnGeN2/GaN heterointerface has inspired novel band-engineered III-N/ZnGeN2 heterostructure-based device designs for applications in high performance optoelectronics. We report on the determination of the valence band offset between metalorganic chemical vapor deposition grown (ZnGe)1−xGa2xN2, for x = 0 and 0.06, and GaN using x-ray photoemission spectroscopy. The valence band of ZnGeN2 was found to lie 1.45–1.65 eV above that of GaN. This result agrees well with the value predicted by first-principles density functional theory calculations using the local density approximation for the potential profile and quasiparticle self-consistent GW calculations of the band edge states relative to the potential. For (ZnGe)0.94Ga0.12N2 the value was determined to be 1.29 eV, ∼10%–20% lower than that of ZnGeN2. The experimental determination of the large band offset between ZnGeN2 and GaN provides promising alternative solutions to address challenges faced with pure III-nitride-based structures and devices. 

The natural band alignment between various II-IV-N$_2$ and III-N and ZnO semiconductors are determined by means of first-principles surface calculations of their electron affinities. While these ignore specific interface dipole formation and strain effects, they provide a first guidance to the construction of heterojunction devices involving this family of materials.