Design and Optimization of a Compact Ultra-Broadband Polarization Beam Splitter for the SCL-Band Based on a Thick Silicon Nitride Platform

Round 1

Reviewer 1 Report

This paper present the design and simulation of Silicon nitride (SiN) based polarisation beam splitters (PBS) as well as a fabrication error tolerance analysis. The proposed PBS architecture is not new and, as stated by the authors, several PBS have already been demonstrated in the SiN platform. However the simulations shows promising performances, so I believe that this manuscript can be published in MDPI considering that some minor issues are adressed : 

1. In order to be fair, In Table 1, an (S) must be added after the PER and IL values of this work to show that it's simulation and not experimental results.

2.  Line 119."...transition of the propagating TE mode...", I believe it's "TM mode".

3.  Paragraph starting at line 141. This paragraph describes the design of the polarization separation region (which is the "heart" of the paper). As mentionned in the manuscript, the most important parameters for the PBS design are deltaneff(TE_strip - TE_slot) and deltaneff(TM_strip - TM_slot) which corresponds to the difference of effective indexes of the fundamental modes of both waveguides (strip and slot) when they are separated (i.e. far from each other). However, for the coupling length discussion (starting after equation (1) in the paper), the considered effective indexes (neffTM1, neffTM2) corresponds to the modes of the coupled system.

I believe that it could be a bit confusing to the reader, please clarify the definition of those effective indexes in the text, or maybe add a reference explaining the coupled mode theory and the beating length equation.

4. Line 165. There is a carriage return in the middle of the sentence.

5. Equation (3), line 185. I believe its "PER_TM" (not PER_TE).

6. Line 294. " the performance of the device is improved 294 by 3.3 dB and 3 dB for TE and TM polarization" I don't understand where does those numbers came from. Is it IL? PER? Please clarify.

7. In figure 8. The vertical axis legend is written delta_t (%); I believe its delta_th (%).

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The authors proposed a compact ultra-broadband polarization beam splitter. This is a potential design idea for photonics on silicon nitride platform. As the paper indicates that it can split light with really high polarization extinction ratio (PER) at the SCL-band. It is a great design for PBS waveguides across a 130 nm span. However, fabrication tolerance analysis also presents the minimum PER of about 22 dB without strict limitations, and deviations for thicknesses of ±16 nm and gaps of ±20 nm. So, fabrication and measurement are necessary for the further research, and more focuses might be needed to slot waveguides embedded in 800 nm thick SiN strips.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

The authors Georgios Patsamanis et al. designed and optimized a compact ultra-broadband polarization beam splitter (PBS) based on the mode coupling between channel waveguide and slot waveguide with different sizes on silicon nitride platform. The key point for obtaining a high performance is to achieve an effective refractive index matching condition between the TM modes in two waveguides while realize a large index difference between the TE and TM modes in the slot waveguide. Although similar PBS structure have been proposed on SOI platform, the author paid great effort to enlarge the wavelength range of the PBS by cascading three PBS with different parameters, and the simulated results have verified their claims. This paper is generally well written and organized. I recommend it to publish in Photonics if the authors can address several small concerns:

1.      In the Introduction part, the authors claim that it is even harder to achieve a high performance of PER, IL, and footprint in SiN waveguide of more than 600 nm thickness, the special advantage for SiN waveguides of more than 600 nm thickness should be elaborated since a smaller thickness will help to realize large index difference between the TE and TM modes.

2.      Table I summarized the key performance metrics of several SiN PBS design, it is suggested to move it to the end of Section 4.

3.      It is not clear enough that why g=250 nm and t=470 nm are chosen, from the result in Figure 2, why can’t one choose g=245 nm and t>470 nm (maybe 480 nm) to satisfy the phase matching condition? It seems that one does not have the necessary to achieve a largest index difference in the normal channel waveguide, the authors did make such a choice in the article.

4.      A typo error exists in Equation (3).

Author Response

Please see the attachment.

Author Response File: Author Response.docx