Assistance Needed for Simulating Dielectric and Mechanical Losses in SAW Devices

Please login with a confirmed email address before reporting spam

Hello everyone,

I am currently simulating a one-port SAW device in COMSOL, focusing on incorporating both dielectric and mechanical losses into the linear materials of Si and Mo, with Mo serving as the IDTs and Si as the substrate where the AlN film (piezoelectric) is deposited. Initially, I used the 'Damping' subnode for mechanical loss, setting it up as an isotropic loss factor, but I am unsure how to simulate dielectric loss. My questions are: (1) Is it correct to use 'Damping' as a form of mechanical loss? (2) Which subnode should be used for dielectric loss in the case of linear elastic material?

The images attached to this annex illustrate the physics nodes I am implementing in my model, along with the damping configuration for the linear elastic material as previously mentioned.

Thank you for your time and assistance.

T.M. Vieira



9 Replies Last Post 3 juil. 2024, 15:07 UTC−4
Edgar J. Kaiser Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 5 days ago 28 juin 2024, 13:11 UTC−4

Hi,

you would typically apply a complex dielectric permittivity if you operate the model in the frequency domain. Check the details in: https://en.wikipedia.org/wiki/Dielectric_loss

Cheers

Edgar

-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Hi, you would typically apply a complex dielectric permittivity if you operate the model in the frequency domain. Check the details in: https://en.wikipedia.org/wiki/Dielectric_loss Cheers Edgar

Please login with a confirmed email address before reporting spam

Posted: 2 days ago 1 juil. 2024, 14:17 UTC−4

Hi,

you would typically apply a complex dielectric permittivity if you operate the model in the frequency domain. Check the details in: https://en.wikipedia.org/wiki/Dielectric_loss

Cheers

Edgar

Hi,

Thank you for your reply. I am operating in the frequency domain; actually, I am replicating a work from a paper where the author operates in both the frequency domain and eigenfrequency. Can the complex dielectric that you refer to also be applied to eigenfrequency? In the settings of Electrostatics > Charge Conservation 1, it is possible to set up Dielectric Loss for the Constitutive Relation D-E. This option asks for the real part and the imaginary part of the complex dielectric. Is this the setting where I can configure the complex dielectric permittivity you are referring to?

Could you inform me if I can use 'Damping' as a form of mechanical loss?

Once more, thank you for your help.

T.M. Vieira

>Hi, > >you would typically apply a complex dielectric permittivity if you operate the model in the frequency domain. Check the details in: https://en.wikipedia.org/wiki/Dielectric_loss > >Cheers > >Edgar Hi, Thank you for your reply. I am operating in the frequency domain; actually, I am replicating a work from a paper where the author operates in both the frequency domain and eigenfrequency. Can the complex dielectric that you refer to also be applied to eigenfrequency? In the settings of Electrostatics > Charge Conservation 1, it is possible to set up Dielectric Loss for the Constitutive Relation D-E. This option asks for the real part and the imaginary part of the complex dielectric. Is this the setting where I can configure the complex dielectric permittivity you are referring to? Could you inform me if I can use 'Damping' as a form of mechanical loss? Once more, thank you for your help. T.M. Vieira

Edgar J. Kaiser Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 2 days ago 1 juil. 2024, 15:39 UTC−4

Yes, you can use damping as a mechanical loss mechanism. The dielectric loss should be a correct approach in electrostatics. Alternatively you can set the complex permittivity in the material properties in the form eps1 + j * eps2. eps1 and eps2 being the real and the imaginary part of the complex permittivity.

Cheers, Edgar

-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Yes, you can use damping as a mechanical loss mechanism. The dielectric loss should be a correct approach in electrostatics. Alternatively you can set the complex permittivity in the material properties in the form eps1 + j * eps2. eps1 and eps2 being the real and the imaginary part of the complex permittivity. Cheers, Edgar

Please login with a confirmed email address before reporting spam

Posted: 1 day ago 2 juil. 2024, 12:55 UTC−4

Yes, you can use damping as a mechanical loss mechanism. The dielectric loss should be a correct approach in electrostatics. Alternatively you can set the complex permittivity in the material properties in the form eps1 + j * eps2. eps1 and eps2 being the real and the imaginary part of the complex permittivity.

Cheers, Edgar

I tried the alternative method of using eps1 + j*eps2, but COMSOL returned the following message: "Undefined value found. Detail: Undefined value found in the equation residual vector." Then, I was wondering if I have to declare 'j' as a variable at some point. Instead of using this method, can I just set up Dielectric Loss in the constitutive relation for Charge Conservation under Electrostatic settings? In this option, it is possible to write the complex permittivity with two separate places for each part of the number (one for real and one for complex).

Thank you for your assistance.

T.M. Vieira

>Yes, you can use damping as a mechanical loss mechanism. The dielectric loss should be a correct approach in electrostatics. >Alternatively you can set the complex permittivity in the material properties in the form eps1 + j * eps2. >eps1 and eps2 being the real and the imaginary part of the complex permittivity. > >Cheers, >Edgar I tried the alternative method of using eps1 + j*eps2, but COMSOL returned the following message: "Undefined value found. Detail: Undefined value found in the equation residual vector." Then, I was wondering if I have to declare 'j' as a variable at some point. Instead of using this method, can I just set up Dielectric Loss in the constitutive relation for Charge Conservation under Electrostatic settings? In this option, it is possible to write the complex permittivity with two separate places for each part of the number (one for real and one for complex). Thank you for your assistance. T.M. Vieira

Edgar J. Kaiser Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 1 day ago 2 juil. 2024, 13:05 UTC−4

Hm, did you literally use "eps1 + j * eps2"? eps1 and eps2 need to be numbers or defined parameters. Sure you can use the constitutive relation. j and i are built in imaginary unity constants in COMSOL. No need to define them.

-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Hm, did you literally use "eps1 + j * eps2"? eps1 and eps2 need to be numbers or defined parameters. Sure you can use the constitutive relation. j and i are built in imaginary unity constants in COMSOL. No need to define them.

Please login with a confirmed email address before reporting spam

Posted: 1 day ago 2 juil. 2024, 13:33 UTC−4

Hm, did you literally use "eps1 + j * eps2"? eps1 and eps2 need to be numbers or defined parameters. Sure you can use the constitutive relation. j and i are built in imaginary unity constants in COMSOL. No need to define them.

I used ((4.5e-11)/8.8e-12)+j*(2e-3) in the space of relative permittivity in properties of the material.

Thank you for your assistance.

T.M. Vieira

>Hm, did you literally use "eps1 + j * eps2"? eps1 and eps2 need to be numbers or defined parameters. Sure you can use the constitutive relation. j and i are built in imaginary unity constants in COMSOL. No need to define them. I used ((4.5e-11)/8.8e-12)+j*(2e-3) in the space of relative permittivity in properties of the material. Thank you for your assistance. T.M. Vieira

Edgar J. Kaiser Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 1 day ago 2 juil. 2024, 14:00 UTC−4
Updated: 1 day ago 2 juil. 2024, 14:00 UTC−4

That should be a valid expression.

-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
That should be a valid expression.

Please login with a confirmed email address before reporting spam

Posted: 4 hours ago 3 juil. 2024, 14:29 UTC−4

That should be a valid expression.

In fact, what happened was that I had incorrectly set up one of the properties of the piezoelectric film. Now it is working. I am still not achieving the same results as in the paper I am trying to replicate, but it is better than before. Thank you for your help. I suspect that I am doing something wrong with the physics of the model. Can you tell just by looking at the picture I sent in the first message if I am forgetting something?

Thank you for your assistance.

T.M. Vieira

>That should be a valid expression. In fact, what happened was that I had incorrectly set up one of the properties of the piezoelectric film. Now it is working. I am still not achieving the same results as in the paper I am trying to replicate, but it is better than before. Thank you for your help. I suspect that I am doing something wrong with the physics of the model. Can you tell just by looking at the picture I sent in the first message if I am forgetting something? Thank you for your assistance. T.M. Vieira

Edgar J. Kaiser Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 3 hours ago 3 juil. 2024, 15:07 UTC−4

No, you need to post the model and describe in more detail what the discrepancies are between your results and the literature

-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
No, you need to post the model and describe in more detail what the discrepancies are between your results and the literature

Reply

Please read the discussion forum rules before posting.

Please log in to post a reply.

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.