Surface reconstruction and energetics for the growth of GaN using atomic-N source

Xiaojie Chen, and W. A. Goddard, III

Abstract

Using the Gaussian Dual Space Density Functional Theory (GDS/DFT), we have investigated the viability of molecular beam epitaxial (MBE) growth of GaN using atomic N source. The c-GaN(100)$_N$ is found to prefer c(2x2) over p(2x1) reconstruction due to the net charge transfer from second layer Ga to top layer N atoms. Energetics relevant to the growth processes are calculated. It is found that ground state atomic N source leads to favour growth against surface-N abstraction, while the excited state (doublet) atomic N source leads to the surface-N abstraction with high probability. It is suggested that for the growth of good quality GaN, one should choose the growth condition such as to enhance the Ga mobility under Ga-rich conditions.