Numerical Method for the Design of Compact Adiabatic Devices with Multiple Parameter Variations

Round 1

Reviewer 1 Report

In this manuscript, a numerical method for designing efficient adiabatic devices with multiple structural parameter variations (NAMSP) was demonstrated. The research about adiabatic taper between the waveguides with different cross sections was important and useful. But the authors need to deal with the following issues and make major revisions before the manuscript can be further considered.

 1.     The novelty should be clarified. In the so-called ‘NAMSP’, the adiabatic taper is divided to several section and for each section of the adiabatic device structure, a linear taper is used. It seems to be common practice for the researchers.

2.     Why does the width of the input waveguide is specified in the mode match point? If we can still get the same conclusion, when the input width is not 3.5um?

3.     Why was the taper of the middle silicon waveguide from 3.5um to 7um divided into two section? Why was the transition width set to be 4um?

Author Response

Response to Reviewer 1

Dear Reviewer 1,

Thank you for your suggestions. We do appreciate the time you have spent and we are working to close this paper with you so that it can be utilized by the Photonics community. We have addressed your points below and incorporated the various suggestions you have given. Reviewing paper is not the most rewarding job, and our thanks to you for spending the time to provide us with valuable feedbacks.  

Comments and Suggestions for Authors

In this manuscript, a numerical method for designing efficient adiabatic devices with multiple structural parameter variations (NAMSP) was demonstrated. The research about adiabatic taper between the waveguides with different cross sections was important and useful. But the authors need to deal with the following issues and make major revisions before the manuscript can be further considered.

Q1. The novelty should be clarified. In the so-called ‘NAMSP’, the adiabatic taper is divided to several section and for each section of the adiabatic device structure, a linear taper is used. It seems to be common practice for the researchers. 

A1. Thanks for the suggestion. Yes, for complex adiabatic device designs (such as adiabatic devices with multiple structural parameter variations in this work), linear tapers are often used when analytical methods are not available. In such cases, the entire structure is connected linearly (such as Ref. [1-2] below), and the entire structure is not divided, as shown in Figure 2 in the manuscript. The drawback of the linear taper design compared to the proposed method in this study is that it leads to a large structure size, as shown in Figure 7 in the manuscript. Thus, the novelty of the research is as follows: A numerical method called “NAMSP” is presented for piecewise minimization of the adiabatic device length by maintaining a constant mode power loss along the device. In addition, to address the design problem of adiabatic devices with multiple structural parameter variations, an efficient domain decomposition scheme is originally introduced into the NAMSP method. The proposed method can be useful for the calculation of compact adiabatic guided-wave shapes for these adiabatic devices with multiple structural parameter variations. 

[1] D. Dai, Y. B. Tang, and J. E. Bowers, “Mode conversion in tapered submicron silicon ridge optical waveguides,” Opt. Express, 20(12), 13425-13439 (2012).

[2]  J. H. Schmid, B. Lamontagne, P. Cheben, A. Delage, S. Janz, A. Densmore, J. Lapointe, E. Post, P. Waldron, and D. X. Xu, “Mode converters for coupling to high aspect ratio silicon-on-insulator channel waveguides,” IEEE Photonics Tech. L., 19(11), 855-857 (2007).  

 

Q2. Why does the width of the input waveguide is specified in the mode match point? If we can still get the same conclusion, when the input width is not 3.5 μm? 

A2. Thanks for the question. When the width of the input waveguide is specified at the mode matching point, the resulting device length will be very compact. And when the input width is not 3.5 μm, the resulting device length will be greater than the length obtained at the mode matching point. This is the reason why the width of the input waveguide is specified at the mode matching point.  

 

Q3. Why was the taper of the middle silicon waveguide from 3.5 μm to 7 μm divided into two section? Why was the transition width set to be 4 μm?

 

A3. Thanks for the question. (1) The choice of the number of sections N to be divided into are iterated after some initial trial simulations before we find a reasonable set of section widths to be used for the final design. In general, the larger the number of sections N, the better the curve of the final structure used for the design, and the better the results determined via the obtained MCTE curve. However, the larger the N, the longer the computational time required. When N is larger than a certain value, there will be diminished returns to improvement. In this case, N around 2 is sufficient and the MCTE curve obtained would approach a limiting curve. Thus, the taper of the middle silicon waveguide is divided into two sections from 3.5 μm to 7 μm; (2) The sensitivity of wide waveguides to width changes is reduced, the width variation near the wide waveguide (7 μm) can be larger, so we set the transition width to 4 μm.

 

Our thanks

We thank you for all the inquisitions and suggestions that make this a better paper. We sincerely thank you for the time you have spent. Hope we have addressed all the suggestions you brought up, and the paper can be published soon so that others can use the method. Your recommendation and support for the publication of this paper would be greatly appreciated.

Author Response File: Author Response.docx

Reviewer 2 Report

In this manuscript, the authors present a numerical method for designing compact adiabatic devices with multiple structural parameter variations. The proposed method can be useful to compute compact adiabatic guided wave shapes for these adiabatic devices with multiple structural parameter variations. It is shown that the efficiency of the adiabatic device designed using the proposed method considerably exceeds that obtained using the linear shape design. This is an interesting research work. The manuscript is recommended for publication in Photonics, provided that the authors address the following questions.

1. What is the width and height of the silicon dioxide in Figure 2? This was not pointed out in the manuscript. Structural parameters need to be considered in the design and simulation.

2. Are the length and MCTE values in Figure 5 unique? Is it possible to change to other parameters?

3. Could the authors explain why the conversion efficiency in Fig. 5 is almost 100% as the waveguide length goes to zero?

4. A 3D structure is recommended for Figure 6, if possible.

Author Response

Response to Reviewer 2

Dear Reviewer 2,

Thank you for your suggestions. We do appreciate the time you have spent and we are working to close this paper with you so that it can be utilized by the Photonics community. We have addressed your points below and incorporated the various suggestions you have given. Reviewing paper is not the most rewarding job, and our thanks to you for spending the time to provide us with valuable feedbacks.  

Comments and Suggestions for Authors

In this manuscript, the authors present a numerical method for designing compact adiabatic devices with multiple structural parameter variations. The proposed method can be useful to compute compact adiabatic guided wave shapes for these adiabatic devices with multiple structural parameter variations. It is shown that the efficiency of the adiabatic device designed using the proposed method considerably exceeds that obtained using the linear shape design. This is an interesting research work. The manuscript is recommended for publication in Photonics, provided that the authors address the following questions.

 

Q1. What is the width and height of the silicon dioxide in Figure 2? This was not pointed out in the manuscript. Structural parameters need to be considered in the design and simulation. 

A1. Thanks for the advice. The width and height of the silicon dioxide in Figure 2 are 8 μm and 1.1 μm, respectively. These parameters have been added in the revised version.  

 

Q2. Are the length and MCTE values in Figure 5 unique? Is it possible to change to other parameters? 

A2. Thanks for the question. The length and MCTE values in Figure 5 are not unique and can be changed to other parameters. We choose a value for the MCTE along the initial slope of the MCTE curve and determine the length for each section that corresponds to the desired mode-connection power loss fraction (MCLF) value. Thus, the value of MCTE is fine as long as it lies on the initial slope of the MCTE curve. The length also varies with the MCTE value.  

 

Q3. Could the authors explain why the conversion efficiency in Figure 5 is almost 100% as the waveguide length goes to zero? 

A3. Thanks. This initial starting value (almost 100%) for the conversion efficiency curve is due to beam power reflection when the waveguide length is zero (L = 0), at which the width of the waveguide (and hence the propagating refractive index of the waveguide mode) is changed abruptly.  

 

Q4. A 3D structure is recommended for Figure 6, if possible. 

A4. Thanks. We have plotted the 3D structure of Figure 6 in the revised version.

 

Our thanks

We thank you for all the inquisitions and suggestions that make this a better paper. We sincerely thank you for the time you have spent. Hope we have addressed all the suggestions you brought up, and the paper can be published soon so that others can use the method. Your recommendation and support for the publication of this paper would be greatly appreciated.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The proposed issues has been replied. I think this manuscript would fit in Applied Optics.