Menu

Discussion Forum

New Discussion
Filter

Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

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.

Optical Metals

Posted Aug 19, 2009, 10:29 p.m. EDT Wave Optics Version 4.1 33 Replies

Asanka Pannipitiya
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Hi,

Does anyone know how to simulate Drude or Loretzian type of metal (gold, silver in Optical frequencies) to see the eigen frequency analysis on plasmonic waveguides...

thanks,
Asanka.
33 Replies Last Post Apr 7, 2013, 3:47 p.m. EDT
Eyal Cohen-Hoshen
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Aug 20, 2009, 2:51 a.m. EDT
Hi,
As I understand the material libraries does not have frequency dependence of the dielectric permitivity.
In this case you should set it by hand for each wavelength you want to measure. This means that you have to run a script with matlab that would set epsilon and lambda for each run. I know that there are FDTD softwares like Optiwave that has the full frequency dependence of epsilon.
Eyal.
Hi, As I understand the material libraries does not have frequency dependence of the dielectric permitivity. In this case you should set it by hand for each wavelength you want to measure. This means that you have to run a script with matlab that would set epsilon and lambda for each run. I know that there are FDTD softwares like Optiwave that has the full frequency dependence of epsilon. Eyal.
Honghui Shen
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Aug 20, 2009, 12:02 p.m. EDT
Hi

I don't know what's the problem you are facing now, you don't know which application, or you don't know how to set the material properties for the subdomain, or something else

I think you can use the mode anlysis application directly.

as for the subdomain setting, if you want to input the full material dispersion into the comsol,
1, you can type the Drude or Loretzian model into the subdomain setting
2, or you can use the "interpolation" to set the material optical constant (first creat a function for the optical constant, then you can use the interpolation )
options->function->'New..'(to define a function, then choose the "interpolation")


Regards
Honghui
Hi I don't know what's the problem you are facing now, you don't know which application, or you don't know how to set the material properties for the subdomain, or something else I think you can use the mode anlysis application directly. as for the subdomain setting, if you want to input the full material dispersion into the comsol, 1, you can type the Drude or Loretzian model into the subdomain setting 2, or you can use the "interpolation" to set the material optical constant (first creat a function for the optical constant, then you can use the interpolation ) options->function->'New..'(to define a function, then choose the "interpolation") Regards Honghui
Yevgeniy Davletshin
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Aug 20, 2009, 1:56 p.m. EDT
Hi Honghui,

Do you know if I need to input Drude model for simulating Mie scattering from gold nanosphere? Or I can just use optical constants from Johnson and Christy data for complex n parameter in Comsol.

Thanks
Hi Honghui, Do you know if I need to input Drude model for simulating Mie scattering from gold nanosphere? Or I can just use optical constants from Johnson and Christy data for complex n parameter in Comsol. Thanks
Honghui Shen
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Aug 20, 2009, 3:32 p.m. EDT
Hi

As far as I know, you can use the optical constant from literature. and this is one of advantages of FEM, namely, using the real optical constant for dispersive materials.

Good luck with your simulations

BTW, does your model in 3D? keep me updated (if you do mind doing this), when you have good results in your simulations. I'm really interested in the plasmonics things. Thank you
Hi As far as I know, you can use the optical constant from literature. and this is one of advantages of FEM, namely, using the real optical constant for dispersive materials. Good luck with your simulations BTW, does your model in 3D? keep me updated (if you do mind doing this), when you have good results in your simulations. I'm really interested in the plasmonics things. Thank you
Yevgeniy Davletshin
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Aug 20, 2009, 3:55 p.m. EDT
Thanks for reply,

I'm trying to setup 3d model to investigate some properties of gold nanoparticles. But at the moment can't get good agreement with Mie solution for spheres.

Will keep you posted.

Thanks for reply, I'm trying to setup 3d model to investigate some properties of gold nanoparticles. But at the moment can't get good agreement with Mie solution for spheres. Will keep you posted.
Honghui Shen
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Aug 20, 2009, 4:28 p.m. EDT


you are trying to investigate the extinction cross section (absorption and scattering) or something else?
you are trying to investigate the extinction cross section (absorption and scattering) or something else?
Yevgeniy Davletshin
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Aug 21, 2009, 9:18 a.m. EDT
in general - yes
in general - yes
John Montoya
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Sep 28, 2009, 2:02 a.m. EDT
Can anyone here please upload your work concerning metals. I'm interested in plasmonic gratings and I'm trying to learn how to use COMSOL for this. If you can show me how to get started, I would greatly appreciate it. Thanks.
Can anyone here please upload your work concerning metals. I'm interested in plasmonic gratings and I'm trying to learn how to use COMSOL for this. If you can show me how to get started, I would greatly appreciate it. Thanks.
Yevgeniy Davletshin
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Sep 30, 2009, 11:02 a.m. EDT
If you want I can upload Mie model
If you want I can upload Mie model
John Montoya
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Sep 30, 2009, 12:04 p.m. EDT
Please upload! Thank you!!
Please upload! Thank you!!
Yevgeniy Davletshin
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Sep 30, 2009, 8:01 p.m. EDT

Please upload! Thank you!!


Done
[QUOTE] Please upload! Thank you!! [/QUOTE] Done
Zhang Jing
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 12, 2009, 9:31 p.m. EDT
You set n=n-j*k in subdomain 4, the 5 nm sphere. I guess k must be a frequency dependent variables, but i can not find where you define the k.
You set n=n-j*k in subdomain 4, the 5 nm sphere. I guess k must be a frequency dependent variables, but i can not find where you define the k.
Yevgeniy Davletshin
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 12, 2009, 10:09 p.m. EDT

You set n=n-j*k in subdomain 4, the 5 nm sphere. I guess k must be a frequency dependent variables, but i can not find where you define the k.


Check original Johnson and Christy paper (1972), n and k are frequency dependent, and both of them defined in the solver parameters. Instead of J&C you can use Palik's "Handbook of optical constants" for example.
[QUOTE] You set n=n-j*k in subdomain 4, the 5 nm sphere. I guess k must be a frequency dependent variables, but i can not find where you define the k. [/QUOTE] Check original Johnson and Christy paper (1972), n and k are frequency dependent, and both of them defined in the solver parameters. Instead of J&C you can use Palik's "Handbook of optical constants" for example.
John Montoya
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 12, 2009, 10:20 p.m. EDT
Question: Let's say we wanted to calculate the absorption spectrum vs. k-vector for this structure. If you use periodic boundary conditions (which would have to be the case in order to save memory), how do you define the source? The plane waves are not in phase with neighboring unit cells when k is not equal to zero.
Question: Let's say we wanted to calculate the absorption spectrum vs. k-vector for this structure. If you use periodic boundary conditions (which would have to be the case in order to save memory), how do you define the source? The plane waves are not in phase with neighboring unit cells when k is not equal to zero.
Yevgeniy Davletshin
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 12, 2009, 10:33 p.m. EDT
John, I don't know.
It's out of scope of mine knowledges.
John, I don't know. It's out of scope of mine knowledges.
Zhang Jing
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 13, 2009, 5:32 a.m. EDT
I think you can use DDA to caculate it. It is an open souce software.

I guess if you simulate it in COMSOL using periodic boundary condition, the memory useage could be hight, especially in 3D modal.
I think you can use DDA to caculate it. It is an open souce software. I guess if you simulate it in COMSOL using periodic boundary condition, the memory useage could be hight, especially in 3D modal.
Zhang Jing
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 13, 2009, 7:07 a.m. EDT
Thanks.

I see that in your model in postprocessing -> General -> Parameter value, there is only one result that lambda0_emw=4.5e-7. The question is that where are the rest results of other wavelength?

I know the "Handbook of optical constants", but I have to simulate Drude type of metal. I set a Drude model in Options -> Function ->New and define a frequency dependent dielectric function "drude", then type "drude" in Subdomain Settings ->Relative permittivity. It works, but the field distribution is the same as the result when I don't set a frequency dependent dielectric function. And I also use a parametric solver and solve for a range of frequency.
Thanks. I see that in your model in postprocessing -> General -> Parameter value, there is only one result that lambda0_emw=4.5e-7. The question is that where are the rest results of other wavelength? I know the "Handbook of optical constants", but I have to simulate Drude type of metal. I set a Drude model in Options -> Function ->New and define a frequency dependent dielectric function "drude", then type "drude" in Subdomain Settings ->Relative permittivity. It works, but the field distribution is the same as the result when I don't set a frequency dependent dielectric function. And I also use a parametric solver and solve for a range of frequency.
Yevgeniy Davletshin
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 15, 2009, 8:46 a.m. EDT
When you will solve it, there will be all wavelengths that are in input parameters for parametric solver.

When you will solve it, there will be all wavelengths that are in input parameters for parametric solver.
Boyang Zhang
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 27, 2009, 2:54 p.m. EDT
I just start to learn comsol for nano-optics, but I am so surprised that it does not provide Optical property of metal.
I just start to learn comsol for nano-optics, but I am so surprised that it does not provide Optical property of metal.
RKR
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 13, 2010, 11:01 a.m. EST
Hi Yevgeniy Davletshin,

Can you just copy me the link where the model is uploaded so that I can have a look at it..

Cheers Ravish
Hi Yevgeniy Davletshin, Can you just copy me the link where the model is uploaded so that I can have a look at it.. Cheers Ravish
RKR
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 14, 2010, 9:41 a.m. EST
Hi...

can you forward me the link where u have uploaded the model...

Thanks
Ravish
Hi... can you forward me the link where u have uploaded the model... Thanks Ravish
Yevgeniy Davletshin
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 18, 2010, 3:44 p.m. EST
Model was deleted. No link exists anymore.
Model was deleted. No link exists anymore.
Haibin Zhu
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 19, 2010, 2:17 a.m. EST

Model was deleted. No link exists anymore.


Hi, I'm concerning about the same problem.
Would you please send me a copy of your model, to my email: zhuhb@sjtu.edu.cn
Thank you!
[QUOTE] Model was deleted. No link exists anymore. [/QUOTE] Hi, I'm concerning about the same problem. Would you please send me a copy of your model, to my email: zhuhb@sjtu.edu.cn Thank you!
RKR
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 19, 2010, 11:15 a.m. EST
Can you upload the model once again .. or can u send me a copy?

Thanks
Ravish
Can you upload the model once again .. or can u send me a copy? Thanks Ravish
RKR
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 19, 2010, 11:15 a.m. EST
Can you upload the model once again .. or can u send me a copy?

Thanks
Ravish
Can you upload the model once again .. or can u send me a copy? Thanks Ravish
RKR
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 19, 2010, 11:15 a.m. EST
Can you upload the model once again .. or can u send me a copy?

Thanks
Ravish
Can you upload the model once again .. or can u send me a copy? Thanks Ravish
RKR
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 19, 2010, 11:15 a.m. EST
Can you upload the model once again .. or can u send me a copy?

Thanks
Ravish
Can you upload the model once again .. or can u send me a copy? Thanks Ravish
RKR
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 19, 2010, 11:15 a.m. EST
Can you upload the model once again .. or can u send me a copy?

Thanks
Ravish
Can you upload the model once again .. or can u send me a copy? Thanks Ravish
RKR
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 19, 2010, 11:18 a.m. EST

Can you upload the model once again .. or can u send me a copy?

Thanks
Ravish


sorry for posting too many times.. thats due to some error in my connection.
[QUOTE] Can you upload the model once again .. or can u send me a copy? Thanks Ravish [/QUOTE] sorry for posting too many times.. thats due to some error in my connection.
Jacob Trevino
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 11, 2010, 9:50 a.m. EST
Hello Everyone,

I know this is an old topic, but it is relevant to my question. I am comfortable in using J&C data, creating a function and interpolating to find appropriate complex index values at various wavelengths. That's fine when I doing scattering simulations over a given spectrum.

My new issue is when trying to find modes or eignfrequencies, similar to the original question of this thread. Since my index data is a function of wavelength, the eigenfrequency solver gives me an error. I guess a simple example would be the step fiber example Comsol works out. How would you find the modes if the outer material had complex/wavelength depended dispersion?

For the eigenfrequency solver, is there a variable for the wavelength?

Anyone had success on this?

Best,
Jacob

Hello Everyone, I know this is an old topic, but it is relevant to my question. I am comfortable in using J&C data, creating a function and interpolating to find appropriate complex index values at various wavelengths. That's fine when I doing scattering simulations over a given spectrum. My new issue is when trying to find modes or eignfrequencies, similar to the original question of this thread. Since my index data is a function of wavelength, the eigenfrequency solver gives me an error. I guess a simple example would be the step fiber example Comsol works out. How would you find the modes if the outer material had complex/wavelength depended dispersion? For the eigenfrequency solver, is there a variable for the wavelength? Anyone had success on this? Best, Jacob
Zhiqi Zhang
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jun 3, 2010, 10:37 a.m. EDT
I meet the same problem, does any one know the answer?
I meet the same problem, does any one know the answer?
Ragini Deore
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Nov 1, 2010, 9:02 p.m. EDT
Hi....have anyone tried with DDA? please let me know if someone can provide help with it?
Hi....have anyone tried with DDA? please let me know if someone can provide help with it?
Constantinos Tsangarides
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Apr 7, 2013, 3:47 p.m. EDT
Hello,

I have the same problem, but in my parameters list I typed (for example):

df=(some constants)*freq

Shouldn't the comsol solve the df for each frequnecy it runs in a range of frequencies in the study ?


Thank you


Hi,
As I understand the material libraries does not have frequency dependence of the dielectric permitivity.
In this case you should set it by hand for each wavelength you want to measure. This means that you have to run a script with matlab that would set epsilon and lambda for each run. I know that there are FDTD softwares like Optiwave that has the full frequency dependence of epsilon.
Eyal.


Hello, I have the same problem, but in my parameters list I typed (for example): df=(some constants)*freq Shouldn't the comsol solve the df for each frequnecy it runs in a range of frequencies in the study ? Thank you [QUOTE] Hi, As I understand the material libraries does not have frequency dependence of the dielectric permitivity. In this case you should set it by hand for each wavelength you want to measure. This means that you have to run a script with matlab that would set epsilon and lambda for each run. I know that there are FDTD softwares like Optiwave that has the full frequency dependence of epsilon. Eyal. [/QUOTE]

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.

Suggested Content