Repository logo
 

Microstructure and reflectance of porous GaN distributed Bragg reflectors on silicon substrates

Published version
Peer-reviewed

Repository DOI


Loading...
Thumbnail Image

Change log

Abstract

Distributed Bragg reflectors (DBRs) based on alternating layers of porous and non-porous GaN have previously been fabricated at the wafer-scale in heteroepitaxial GaN layers grown on sapphire substrates. Porosification is achieved via the electrochemical etching of highly Si-doped layers, and the etchant accesses the n+-GaN layers through nanoscale channels arising at threading dislocations that are ubiquitous in the heteroepitaxial growth process. Here, we show that the same process applies to GaN multilayer structures grown on silicon substrates. The reflectance of the resulting DBRs depends on the voltage at which the porosification process is carried out. Etching at higher voltages yields higher porosities. However, while an increase in porosity is theoretically expected to lead to peak reflectance, in practice, the highest reflectance is achieved at a moderate etching voltage because etching at higher voltages leads to pore formation in the nominally non-porous layers, pore coarsening in the porous layers, and in the worst cases layer collapse. We also find that at the high threading dislocation densities present in these samples, not all dislocations participate in the etching process at low and moderate etching voltages. However, the number of dislocations involved in the process increases with etching voltage.

Description

Journal Title

Journal of Applied Physics

Conference Name

Journal ISSN

0021-8979
1089-7550

Volume Title

Publisher

American Institute of Physics

Rights and licensing

Except where otherwised noted, this item's license is described as Attribution 4.0 International
Sponsorship
Engineering and Physical Sciences Research Council (EP/R03480X/1)
EPSRC (EP/W03557X/1)
Engineering and Physical Sciences Research Council (EP/N509620/1)
Engineering and Physical Sciences Research Council (2278538)
EPSRC (EP/R513180/1)
This research was supported by the EPSRC under EP/R03480X/1, EP/W03557X/1 and EP/X015300/1. Rachel Oliver would like to acknowledge funding from the Royal Academy of Engineering under the Chairs in Emerging Technologies Scheme, which is sponsored by the Department for Science, Innovation and Technology (DSIT). We thank Clifford McAleese for his work on the transfer matrix model.