Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Mode Analysis of silicon waveguide

Please login with a confirmed email address before reporting spam

Hello,

I'am doing a simulation with RF module. It's about the mode analysis of a rectangular silicon waveguide.

The waveguide is a rectangular parallelepiped of silicon with the width 0.8 um, the height 0.6 um and the length 20um. And it's attached on a SiO2 substrate with the same length of 20um. And the medium above the waveguide is Air.

All my calculations are based on refractive index, so the index is the only parameter what I defined for the 3 mediums(Si, Air, SiO2).

To propagate the monochromatic light of 1550nm (1.935e14 Hz ), I've put one extremity of the length of Silicon waveguide as "excitation port"(type "numeric", input power:1watt); and the other side as "listener port". For the frequency domain analysis, I put 50 points in the range of (1.9e14 Hz, 2.0e14 Hz).

For the result of the "Boundary Mode Analysis", I can obtain the solutions of "effective mode index", but as for the distribution of intensity of electric field, it's 0 everywhere.

So my question is why I can't obtain a good distribution of electric field, because 0 everywhere seems abnormal theoretically (at least I think there should be something instead of 0). Maybe there are some nessesary parameters that I haven't defined?

Waiting sincerely for your help, I'am really a "rookie" for COMSOL.

Best regards,
Shuo WANG


10 Replies Last Post 09.10.2012, 06:04 GMT-4
COMSOL Moderator

Hello Shuo Wang

Your Discussion has gone 30 days without a reply. If you still need help with COMSOL and have an on-subscription license, please visit our Support Center for help.

If you do not hold an on-subscription license, you may find an answer in another Discussion or in the Knowledge Base.


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 28.09.2011, 04:55 GMT-4
Hi,

I am having a similar problem in optical fiber, did you solve the problem by any chance?

I think a lot still.

I set up the link of my thread and attach a jpg file.

www.comsol.com/community/forums/general/thread/22716/

and I attached a mph file(ver 4.0a) there.




Hi, I am having a similar problem in optical fiber, did you solve the problem by any chance? I think a lot still. I set up the link of my thread and attach a jpg file. http://www.comsol.com/community/forums/general/thread/22716/ and I attached a mph file(ver 4.0a) there.


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 28.09.2011, 07:04 GMT-4
I have a similar problem, but my waveguide stops after the entry to the port. if i disable the port and i am using the scattering boundary condition i have waveguide through the entire fiber but i can not establish the desired mode.
I have a similar problem, but my waveguide stops after the entry to the port. if i disable the port and i am using the scattering boundary condition i have waveguide through the entire fiber but i can not establish the desired mode.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 28.09.2011, 07:39 GMT-4
I thought that i had to use the boundary condition, too. But I attempted that, the result did not come out well.

If we select the Refractive Index in the Dispacement Field of the Wave Equation-Electric, because COMSOL processing algorithms solve the total reflection between core and clad of fiber internally, i think that we don't have to create the Scattering Boundary Condition.

Should we use the Perfectly Matched Layers by any chance?
(Even, I put some Electric Field directly into the core of the optical fiber. but it was also wrong to do that .)

However, I cannot assure. I would like to listen to a second opinion.


I appreciate your interest,

Yong-soo Kim.
I thought that i had to use the boundary condition, too. But I attempted that, the result did not come out well. If we select the Refractive Index in the Dispacement Field of the Wave Equation-Electric, because COMSOL processing algorithms solve the total reflection between core and clad of fiber internally, i think that we don't have to create the Scattering Boundary Condition. Should we use the Perfectly Matched Layers by any chance? (Even, I put some Electric Field directly into the core of the optical fiber. but it was also wrong to do that .) However, I cannot assure. I would like to listen to a second opinion. I appreciate your interest, Yong-soo Kim.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 09.10.2011, 08:26 GMT-4
What kind of error detected by the system?
If there is "out of memory", you'd better check the RAM capacity and the operating system of your PC.
With a PC of 64-bit system, it shall never happen the "out of memory" error.
A 64-bit operating system and 12GB (or even more) RAM is capable to conduct a 3D model simulation.

For the boundary conditions, you could activate "Scattering Boundary Conditions" and adopt the defaut values, then assign this "Scattering Boundary Conditions" to all the outer boundaries.

By the way, it's nessesary to make your "mesh" as simple as possible, because the complexity of mesh lead consequently to complecated calculations of the machine and a great demand of RAM capacity. You'd better simplify the mesh as long as it won't infulence the correctness of the result.

Hope that could help you.
What kind of error detected by the system? If there is "out of memory", you'd better check the RAM capacity and the operating system of your PC. With a PC of 64-bit system, it shall never happen the "out of memory" error. A 64-bit operating system and 12GB (or even more) RAM is capable to conduct a 3D model simulation. For the boundary conditions, you could activate "Scattering Boundary Conditions" and adopt the defaut values, then assign this "Scattering Boundary Conditions" to all the outer boundaries. By the way, it's nessesary to make your "mesh" as simple as possible, because the complexity of mesh lead consequently to complecated calculations of the machine and a great demand of RAM capacity. You'd better simplify the mesh as long as it won't infulence the correctness of the result. Hope that could help you.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 12.10.2011, 23:48 GMT-4
Hi,

You state the key point for me!

Thanks! I will try that.

Hi, You state the key point for me! Thanks! I will try that.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 12.04.2012, 11:18 GMT-4
Hi Shuo,

I have the same problem as you had. I have challenged with my structure for several days, but I have not yet found how to deal with.
I'll appreciate it if you could share your structure (or send it to my email) if you have solved it.

ahajebi2000@gmail.com

Thank you in advance.
Akram
Hi Shuo, I have the same problem as you had. I have challenged with my structure for several days, but I have not yet found how to deal with. I'll appreciate it if you could share your structure (or send it to my email) if you have solved it. ahajebi2000@gmail.com Thank you in advance. Akram

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 04.07.2012, 16:55 GMT-4
Hello All,

I am facing the same problem. No Electric Field is transmitted, it is always zero.
I have a Crystal fiber where I intend to get the input and output power.
I have set an input port as numeric with 1 W and the output port as numeric, but without excitation.

Also my model has a layer of PML outside.

Does any one has an idea about what is the error?

Thanks in advance.

Felipe C. Salgado
Hello All, I am facing the same problem. No Electric Field is transmitted, it is always zero. I have a Crystal fiber where I intend to get the input and output power. I have set an input port as numeric with 1 W and the output port as numeric, but without excitation. Also my model has a layer of PML outside. Does any one has an idea about what is the error? Thanks in advance. Felipe C. Salgado

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 04.07.2012, 17:00 GMT-4
Hi,

My problem was solved.
Please share your job, maybe by looking at your simulation I can find the problem.
Hi, My problem was solved. Please share your job, maybe by looking at your simulation I can find the problem.

Clenilson Rodrigues da Silveira

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 23.07.2012, 09:47 GMT-4
Hi everyone! I have the same problem with my structure based on optical fiber. I got the effective index modes but no electric field distributuion. I realized that it is a commom problem. Please, I'd like to get information about what I should to do to solve this problem?

Thanks in advance!
Hi everyone! I have the same problem with my structure based on optical fiber. I got the effective index modes but no electric field distributuion. I realized that it is a commom problem. Please, I'd like to get information about what I should to do to solve this problem? Thanks in advance!

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 09.10.2012, 06:04 GMT-4

Hi Shuo,

I have the same problem as you had. I have challenged with my structure for several days, but I have not yet found how to deal with.
I'll appreciate it if you could share your structure (or send it to my email) if you have solved it.

ahajebi2000@gmail.com

Thank you in advance.
Akram


Hey Akram,

I'm also the beginner of this simulation software, struggling this case for a long while. And I'm curious about whether you mind or not sending your case to me? Thank you so much for your kindness and great patience.

denilson12001@gmail.com

Best,

Mark
[QUOTE] Hi Shuo, I have the same problem as you had. I have challenged with my structure for several days, but I have not yet found how to deal with. I'll appreciate it if you could share your structure (or send it to my email) if you have solved it. ahajebi2000@gmail.com Thank you in advance. Akram [/QUOTE] Hey Akram, I'm also the beginner of this simulation software, struggling this case for a long while. And I'm curious about whether you mind or not sending your case to me? Thank you so much for your kindness and great patience. denilson12001@gmail.com Best, Mark

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.