Near- and Far-Field Excitation of Topological Plasmonic Metasurfaces

Round 1

Reviewer 1 Report

The work is a rigorous theoretical study, but I have a number of comments before publishing it in the journal:

 

- The authors wrote: “Despite 2 of 15 32 both having a trivial Z2 index, the phases of the breathing honeycomb are topologically distinct”.  How this correlates with the calculation of spin Chern number in the original paper by Wu and Hu [PRL 114, 223901 (2015)]? They do show nonzero spin Chern number, the same is obtained in later works. Please, comment on that.

 

- The authors wrote: “there is no comprehensive theoretical study of the methods for exciting pseudospin edge modes and in particular, the necessary conditions for exciting unidirectional modes”.  However, it's a bit strange statement, because it is widely known that one or another topological modes can be excited by the circularly polarized dipole which matches the polarization of the respective mode.

 

- Equation (6) is a nonlinear eigenvalue problem. Unknown \omega enters a nonlinear equation. How do the authors solve it? More detailed description is needed.

 

- The authors wrote: “We note that the ordering in this plasmonic metasurface is opposite to the photonic crystal 86 due to the metallic nature of the NPs”. Does it mean that now shrunken lattice becomes topologically non-trivial? Please, comment on that.

 

- The authors imply that exciting the system with point sources with circularly-polarized magnetic fields is something unique. However, this idea is not so new and it was exploited in a series of works, including e.g. [37].

 

- Generally, the manuscript looks like the follow-up of the authors’ previous study [https://pubs.acs.org/doi/10.1021/acsphotonics.9b01192]. Can the authors stress novelty more clearly?    

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The manuscript studies numerically the directional excitation of topological edge modes in a topological photonic waveguide formed by an interface between shrunken and expanded honeycomb arrays of plasmonic nanoparticles. The simulations reveal how the directionality of the excited edge modes is sensitive to the position of the source. Namely, for a near field excitation the directionality is relatively insensitive to the source height. On the other hand, the directionality is sensitive to in-plane shifts of the source for both near and far-field excitations. This work is timely and of interest to researchers working on topological photonic waveguides. The introduction clearly places the work in context and the text and figures generally read well. Therefore I think the manuscript is suitable for publication in Photonics. I have some minor suggestions which should be addressed prior to publication:

1. Is is not entirely clear what k_{\parallel} is used for plotting Fig. 1(e). Is it the solid vertical lines in (c,d)? The orange line in (d) is very hard to see against the orange background. The colours and caption should be amended.

2. According to the caption of Fig. 1(f), the losses have been decreased to increase the visibility of the edge states. It would be useful to include a similar plot for realistic losses (e.g. in the Appendix) so that the reader can see whether this edge state dispersion would in principle be visible using this design, as the losses seem to exceed the size of the topological band gap.

3. On a related note, above Eq. (10) authors state that the losses at the interface are set to zero "in order to test the directionality behaviour". I think this potentially significant point needs to be clarified, as one wonders whether the strong losses here may affect the directionality. According to many recent works, losses can have a big effect on topological phases and edge states. Therefore this should be checked, and additional plots included if the losses do lead to qualitatively different behaviour. Of course, the present results remain valid for lossless topological wave systems such as dielectric structures.

4. It is a little confusing that various names are used to describe the frequency ranges of interest. For example, Sec. 3 describes the inversions of the bulk bands. Then in the following Sections "upper band" and "lower band" are used to refer to the two branches of the edge states which lie in the bulk band gap. This might be confused with the bulk bands.

Author Response

Please see the attachment

Author Response File: Author Response.pdf