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.

Piezoelectric Coordinate System For SAW, BAW Lamb waves Simulation

Please login with a confirmed email address before reporting spam

Hi Everyone,

I have a hard time understanding piezoelectric coordinate system in general and for MEMS applications in specific

Right now i am trying to model SAW simulation

I attende one of Comsol webinar regarding SAW and the lecturer used a YZ lithium niobate since it is commonly used. Unfortunately since there was lots of questions in the webinar, my questions were not addressed so I am hoping

The thing that I do not understand is that (He specified a rotated coordinate system for the lithium niobate as follow = " For YZ-LiNbO3 the rotations are { alpha = -pi/2; beta=pi/2; gama=0} and said the YZ- means y surface and z propagation)

If I understood correct from reading about coordinate system rotation is that we are trying to align the crystal orientation to the global orientation by setting Euler angles with respect to the global

If that was right then my questions in my mind are:

  1. There are other lithium niobate like 128yx and others. What is the difference between each one of them in Comsol? Is it just propogation direction and suface difference ?
  2. What happens if we did not set up any coordinate system and keep it to the global system ? What is the initial coordinate system of the piezoelectric materials ?
  3. Where could I get the Euler angles of the piezoelectric marterials?

  4. If I want to simulate SAW and Lamb waves of the same material, do the coordinate system change to generate different waves or is that the coordinate system is not relates to the waves but related to the crystal orientation of the materials and the waves are generated based on the model design and the electrode excitations?

Thank You

Regards, Abdul


2 Replies Last Post 11.02.2018, 23:18 GMT-5

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 11.02.2018, 13:00 GMT-5

Piezoelectric materials are anisotropic. The wave velocities, etc. vary depending on the direction of propagation. So surface wave velocities depend on the surface and the direction of propagation on that surface.

COMSOL has built-in parameters for some piezoelectric materials. These parameters are specified in the global coordinate system. You can EITHER define your domain so that waves are propagating on the right surface in the right direction, OR you can create a new rotated coordinate system for the material to achieve the same result.

SOME orientations do not require a 3D solution. The YZ-lithium niobate direction can be solved in 2D. In 2D you CAN choose the material orientation (look under Solid Mechanics > Piezoelectric Material 1 > Coordinate system).

For other materials and propagation directions of interest- particularly some rotated substrates- it is necessary to to use 3D. As I noted, you can either specify a rotation for the domain or use a rotated coordinate system for the material.

The Euler angles to specify rotated coordinate systems are defined in a figure within COMSOL. I recommend reading carefully and understanding the crystallogaphy of the material you are interested in. Then work out the necessary rotations AND verify that your calculated velocities are consistent with the literature.

D.W. Greve DWGreve Consulting

Piezoelectric materials are anisotropic. The wave velocities, etc. vary depending on the direction of propagation. So surface wave velocities depend on the surface and the direction of propagation on that surface. COMSOL has built-in parameters for some piezoelectric materials. These parameters are specified in the global coordinate system. You can EITHER define your domain so that waves are propagating on the right surface in the right direction, OR you can create a new rotated coordinate system for the material to achieve the same result. SOME orientations do not require a 3D solution. The YZ-lithium niobate direction can be solved in 2D. In 2D you CAN choose the material orientation (look under Solid Mechanics > Piezoelectric Material 1 > Coordinate system). For other materials and propagation directions of interest- particularly some rotated substrates- it is necessary to to use 3D. As I noted, you can either specify a rotation for the domain or use a rotated coordinate system for the material. The Euler angles to specify rotated coordinate systems are defined in a figure within COMSOL. I recommend reading carefully and understanding the crystallogaphy of the material you are interested in. Then work out the necessary rotations AND verify that your calculated velocities are consistent with the literature. D.W. Greve DWGreve Consulting

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 11.02.2018, 23:18 GMT-5
Updated: 7 years ago 12.02.2018, 12:22 GMT-5

Hi David,

Thank You Very Much for your webinar, it was very helpful for my research and thank you for having the time to reply to my post I really appreciate it

I have some questions in my mind

  1. If I understood you right, I still can see the mode and the eigenfrequency of the saw even if I kept the coordinate system to the Global BUT my simulation would have incorrect frequency and therefore my velocity value would be totally off completely wrong, Is that right? (I did that to compare with the tutorial and it changed from 3.48E8 to 3.79E8). But if what I saw by keeping the orientation to the global orientation is indeed SAW mode, does that still mean I could generate and see the modes of SAW but with wrong results?

  2. In Comsol application library, there is a model for SAW gas sensor in which 2D model was used and the material specified was YZ-Cut lithium niobate and the material was not taken from Comsol library but it was created from a blank material, and in the coordinate system it was kept as the default in 2D simulation which is XZ-Plane eventhough there is an option for YZ-Plane, why it was not changed to YZ-Plane?

  3. If I have a piezoelectric material with specified parameter getting from a book or literature and I don't know the crystal cut orientation like ZnO and AlN, do you suggest me to do 2D simulation to verify the crystal cut then redo it in 3D simulation and then rotating it using Euler angles?

  4. Can I consider SAW as a verification method to have the right orientation for the material in order to simulate BAW and Lamb waves? Or do each one have different orientation different Euler Angles because SAW propogate on the surface and BAW on the thickness and Lamb on the lateral?

Thank You for your valuable time to answer my questions

Best Regards, Abdulrahman Alsolami

Hi David, Thank You Very Much for your webinar, it was very helpful for my research and thank you for having the time to reply to my post I really appreciate it I have some questions in my mind 1. If I understood you right, I still can see the mode and the eigenfrequency of the saw even if I kept the coordinate system to the Global BUT my simulation would have incorrect frequency and therefore my velocity value would be totally off completely wrong, Is that right? (I did that to compare with the tutorial and it changed from 3.48E8 to 3.79E8). But if what I saw by keeping the orientation to the global orientation is indeed SAW mode, does that still mean I could generate and see the modes of SAW but with wrong results? 2. In Comsol application library, there is a model for SAW gas sensor in which 2D model was used and the material specified was YZ-Cut lithium niobate and the material was not taken from Comsol library but it was created from a blank material, and in the coordinate system it was kept as the default in 2D simulation which is XZ-Plane eventhough there is an option for YZ-Plane, why it was not changed to YZ-Plane? 3. If I have a piezoelectric material with specified parameter getting from a book or literature and I don't know the crystal cut orientation like ZnO and AlN, do you suggest me to do 2D simulation to verify the crystal cut then redo it in 3D simulation and then rotating it using Euler angles? 4. Can I consider SAW as a verification method to have the right orientation for the material in order to simulate BAW and Lamb waves? Or do each one have different orientation different Euler Angles because SAW propogate on the surface and BAW on the thickness and Lamb on the lateral? Thank You for your valuable time to answer my questions Best Regards, Abdulrahman Alsolami

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.