Grating Coupler Design for Low-Cost Fabrication in Amorphous Silicon Photonic Integrated Circuits
, Alessandro Fantoni 1,2,*, João Costa 1,2 and Manuela Vieira 1,2,3
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsTo authors:
In the submitted manuscript, the authors propose fully-etched grating couplers with fewer fabrication steps while maintaining high performance. The manuscript provides detailed introductions and descriptions of the designed grating couplers. To enhance its contribution, the authors are kindly requested to address the following concerns:
1. Sections 2.1 to 2.4 read more like a review paper due to the extensive detail on different couplers and grating couplers. However, the introduction section fails to provide detailed information on other grating couplers and the rationale for choosing hydrogenated amorphous silicon. The main advantages of this method compared to others should be highlighted. Additionally, the paper’s overall organization, including the four different types of grating coupler designs and key results, should be outlined in the introduction section. Reorganize the introduction and sections 2.1 to 2.4 for better logical flow.
2. In figure 4 (c), the draw of the tapered lensed optical fiber is not clear.
3. Line 130, “the 1st and 3rd methods…” should specify "as shown in figure 4" for clarity.
4. All instances of “Figure” after a comma should be lowercase, i.e., “figure.”
5. Line 186, the sentence “in this case,… the transverse electric (TE) mode” is confusing and should be rephrased for clarity.
6. Line 205, it should be “equation (2)”, instead of just “(2)”.
7. Lines 211 to 213, the authors should provide references for the data on coupling efficiency under -9 dB and a coupling angle of 16.5°.
8. Line 307, the authors should give a more detailed description of the computational algorithm and the rationale for choosing the pseudo-code algorithm.
9. In figure 14, the meanings of the red and green color should be clarified.
10. Line 342, the authors should provide references for the data on modest efficiency over -10 dB or -9dB.
11. In section 2.9, the author should provide details on the algorithm used for the random distribution overlapped grating design. Why does the authors conclude AI algorithm would perform better in the conculsion section.
12. Figures 19, 21, 24, 27, 30, and 33 should include explanations for the abscissa label “cT” in the text.
13. Line 681, a comma is missing.
14. All the figures should be centered.
Author Response
We would like to thank the reviewer for their time and commitment in elaborating a constructive set of comments and suggestions. We have tried to address all the points that was raised. The new added text is highlighted in the manuscript.
Answers for Reviewer 1
- Sections 2.1 to 2.4 read more like a review paper due to the extensive detail on different couplers and grating couplers. However, the introduction section fails to provide detailed information on other grating couplers and the rationale for choosing hydrogenated amorphous silicon. The main advantages of this method compared to others should be highlighted. Additionally, the paper’s overall organization, including the four different types of grating coupler designs and key results, should be outlined in the introduction section. Reorganize the introduction and sections 2.1 to 2.4 for better logical flow.
Answer 1: The main advantages are the absence of a back-reflector or buried layer for increased reflectivity and the ability to use few lithographic steps. Hydrogenated amorphous silicon can be deposited at low temperatures by means of PECVD and HWCVD, the fabrication process is compatible with CMOS technology. We have rephrased the text and made improvements to make this more clear in paragraphs 2, 5 and 6 of the Introduction, lines 43 to 45, 84 to 92 and 95 to 113. We have moved the review section for a dedicated chapter (Section 2. Brief Review of Light Coupling Methods), with a dedicated grating coupler subchapter (Section 2.1. Grating Couplers).
- In figure 4 (c), the draw of the tapered lensed optical fiber is not clear.
Answer 2: The figure, which now is figure 2c, depicts an optical fiber with a tapered tip, the caption was corrected.
- Line 130, “the 1stand 3rd methods…” should specify "as shown in figure 4" for clarity.
Answer 3: The text of figure 2 (previous figure 4) was improved as suggested by the reviewer.
- All instances of “Figure” after a comma should be lowercase, i.e., “figure.”
Answer 4: The text was improved as suggested by the reviewer.
- Line 186, the sentence “in this case,… the transverse electric (TE) mode” is confusing and should be rephrased for clarity.
Answer 5: The text was improved as suggested by the reviewer.
- Line 205, it should be “equation (2)”, instead of just “(2)”.
Answer 6: The text was improved as suggested by the reviewer.
- Lines 211 to 213, the authors should provide references for the data on coupling efficiency under -9 dB and a coupling angle of 16.5°.
Answer 7: A mention to RSoft software package (Synopsys, Inc), FullWAVE – 2D-FDTD implementation, was added in the 10th paragraph of section 3.3 (previous section 2.3), lines 251 to 253. Two references of uniform grating couplers were added to compared with the results obtained in simulation, lines 253 to 258 of section 3.3 (previous section 2.3). Table 4 of section 5 (previous section 4) was also updated with the most relevant results for these two references.
- Line 307, the authors should give a more detailed description of the computational algorithm and the rationale for choosing the pseudo-code algorithm.
Answer 8: The pseudo-code algorithm is used to calculate the grating period and fill factor for each grating segment (etch and unetched “parts”), considering the effective refractive index variation through the grating coupler’s length, thus facilitating grating coupler representation in CAD. New text was added in lines 309 to 311, of section 3.5 and lines 351 to 353 of section 3.6.
- In figure 14, the meanings of the red and green color should be clarified.
Answer 9: The caption of figure 15 (previous figure 14) was improved as suggested.
- Line 342, the authors should provide references for the data on modest efficiency over -10 dB or -9dB.
Answer 10: These values were obtained from the 2D-FDTD simulations. We made clearer in the text that these results were obtained by simulation, section 3.8, lines 390 to 392. A reference was included in the same section, lines 392 to 394.
- In section 2.9, the author should provide details on the algorithm used for the random distribution overlapped grating design. Why does the authors conclude AI algorithm would perform better in the conculsion section.
Answer 11: Changed reference in the 4th chapter of section 5 (previous section 4), lines 725 to 727, to: “We were unable to achieve acceptable results from pseudo-random distributions, nevertheless these could be the basis for more sophisticated optimization, such as for example genetic algorithms, by creating a seed samples.”.
- Figures 19, 21, 24, 27, 30, and 33 should include explanations for the abscissa label “cT” in the text.
Answer 12: The abscissa label “cT” corresponds to a time-related unit which is given as the product between the velocity of light in vacuum () and time (). Since the velocity of light is constant, this would represent the space travelled by light for a given time interval, in units of micrometer (μm). This is a common unit in numerical simulation software. We have included this explanation in the text of the manuscript to clarify the meaning of “cT”. The new text can be found in lines 469 to 473, of section 4.3.
- Line 681, a comma is missing.
Answer 13: The text was improved as suggested by the reviewer.
- All the figures should be centered.
Answer 14: We didn’t want to change the template parameters. We will be asking for input by the editorial office for this task.
Reviewer 2 Report
Comments and Suggestions for AuthorsDear Editor,
This study focuses on fully-etched grating couplers without bottom reflectors, made from a-Si:H. The manuscript should be polished based on the below comments. So, I recommend the major revision in this round.
1. There are other methods for light coupling, including the use of prisms. It is better to discuss this issue in the introduction of the article.
2. In what wavelength range are the results obtained in this research valid?
3. Compare the loss curves of SiO2 and SiNx.
4. It is necessary to examine the obtained results for the angular error of 1, 2, and 5 degrees in the vertical etching of gratings.
5. Draw the real and imaginary components of the waveguide permittivity and discuss their changes according to the structural characteristics of the gratings.
6. In Table 4, why is the coupling efficiency value improved compared to other works? Discuss physically.
7. Calculate and draw the curve related to the light confinement in the waveguide.
Kind regards
Author Response
We would like to thank the reviewer for their time and commitment in elaborating a constructive set of comments and suggestions. We have tried to address all the points that was raised. The new added text is highlighted in the manuscript.
Answers for Reviewer 2
- There are other methods for light coupling, including the use of prisms. It is better to discuss this issue in the introduction of the article.
Answer 1: Since the material is a-Si:H, the refractive index is very high, around 3.5, for effective light coupling the prism should have a refractive index higher than the waveguide’s core material. Prisms with such high refractive index and low extinction coefficient are not commercially available at the wavelength of interest (1550 nm). To better elucidate this matter, we have included new text in the Introduction section, lines 45 to 53.
- In what wavelength range are the results obtained in this research valid?
Answer 2: The results are valid for a wavelength of 1550 nm, to make it clearer we also made a new reference to the wavelength on the first paragraph of section 5 (previous section 4), lines 694 and 695, and in the caption of table 4, lines 698 and 699.
- Compare the loss curves of SiO2 and SiNx.
Answer 3: Since SiNx (silicon nitride) is not used in the proposed devices and SiO2 (silica) is the material which composes the substrate and optical fiber, not the core of the waveguides. While SiNx is an interesting candidate for photonic integrated circuits, this analysis would be outside the scope of this manuscript, that is exclusively focused on a-Si:H structrues.
- It is necessary to examine the obtained results for the angular error of 1, 2, and 5 degrees in the vertical etching of gratings.
Answer 4: This is a very pertinent observation, since the fabrication imperfections may have a significant impact on the performance of the photonic devices. In addition to the vertical angular deviation, it would be also interesting to include an analysis regarding the effects of wall roughness and mask alignment errors. This would be a subject foreseen for a subsequent study, however the focus of this manuscript is the comparison of the efficiency of different grating coupler topologies, independently of the fabrication conditions. While at present time our study is limited to a simulation approach, in the next future we plan to fabricate the gratings, and at that time we will produce a detailed analysis about the differences that can be expected from the simulation results and which parameters are most critical for this step.
- Draw the real and imaginary components of the waveguide permittivity and discuss their changes according to the structural characteristics of the gratings.
Answer 5: The permittivity of the waveguide core and cladding (substrate) does not change with the structural characteristics of the gratings, so we assume the reviewer is asking about the plot of the permittivity variation of the waveguide material versus the wavelength. A new plot was included in figure 5 (previous figure 2).
- In Table 4, why is the coupling efficiency value improved compared to other works? Discuss physically.
Answer 6: The reason behind the improved performance is the adjustment of the fill-factor (and period in apodized designs), in order to shape the diffracted field to the coupling profile of the fundamental mode of the optical fiber. To make this approach clearer, it is now mentioned in section 5 (previous section 4), paragraph 3, lines 713 to 717.
Calculate and draw the curve related to the light confinement in the waveguide.
Answer 7: A new plot (figure 17) was added in section 4.1 (previous section 3.1) of the manuscript as suggested. The waveguide light confinement curve is referenced and discussed in lines 428 to 436, of chapter 4.1.
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsDear editor,
The comments have been addressed so the present form of the revised manuscript can be published.
Kind regards
