In the above paper
[1], there was a minor error in the implementation of the transfer matrix method used to calculate the complex reflection coefficient
of the nanorod/GaN/sapphire trilayer. The formulas provided by Hecht
[2] use the convention, where the phase of the electromagnetic fields is assumed to vary as e
as opposed to the convention used in the rest of the work where the phase varies as e
. While this inconsistency has affected the results reported in
[1, Figs. 5–9], the authors believe the conclusions drawn from these data are unchanged.