*INITIAL CONDITIONS, TYPE=STRESS error

[boreilly]

Hi, I am using GMSH and Calculix to simulate the lattice mismatch stress between Silicon & Silicon Germanium.
The difference in lattices is 4.2%. I am trying to apply this as an initial stress at the interface between the 2 materials.
I am trying to use the Initial Conditions command.

*INITIAL CONDITIONS, TYPE=STRESS
interface,1,4.2%

I get error reading *INITIAL CONDITIONS

Any advice/guidance appreciated.

Regards,
Barry.

[jmaustpc]

Any advice/guidance appreciated.

Please post your version data in the manner explained in the "forum rules " link above.

[boreilly]

Apologies, I do not have FreeCAD, I am using Calculix v0.32.

Thanks,
Barry.

[UR_]

I do not have FreeCAD

Bad news first

*INITIAL CONDITIONS, TYPE=STRESS
interface,1,4.2%

This didn't fulfill ccx's syntax requirements.

Please see docs here: (page 404)
http://www.dhondt.de/ccx_2.13.pdf

AFAIU, you shouldn't apply an initial stress, but an initial strain (see page 405)

And of course ccx doesn't work with percentages

[HarryvL]

Hi, I am using GMSH and Calculix to simulate the lattice mismatch stress between Silicon & Silicon Germanium.
The difference in lattices is 4.2%. I am trying to apply this as an initial stress at the interface between the 2 materials.
I am trying to use the Initial Conditions command.

*INITIAL CONDITIONS, TYPE=STRESS
interface,1,4.2%

I get error reading *INITIAL CONDITIONS

Any advice/guidance appreciated.

Regards,
Barry.

Barry, i am not quite sure what mechanical problem you are trying to solve. What does the difference of 4.2% mean? Between what and what? Could you add a sketch of the problem you are trying to analyze? Is it a layered arrangement of materials with different temperature expansion characteristics? Harry

[NormandC]

This has nothing to do with FreeCAD.

Hence, it should not be in the "Help on using FreeCAD" forum. Obviously.

[boreilly]

Apologies again for using the wrong forum.

Hi Harry,
Thank for the reply I appreciate your time,

Please find the following links to understand the problem I am trying to simulate using FEA.

http://cmos.fe.up.pt/wiki/public/tutori ... ionprocess Please see figure 13.

http://www-inst.eecs.berkeley.edu/~ee29 ... ture10.pdf Please see page 2 Substrate Induced Strain.

By growing a layer of silicon on a silicon Germanium base, a biaxial tensile stress is induced in the upper silicon layer. This results from the difference in lattice constant of the two layers at the atomic level.
During fabrication the silicon is forced to take the larger lattice constant of the Silicon Germanium layer. The resulting stress results in changes to the material properties of the silicon leading to enhanced performance of electronic Mosfets by increasing current flow.

I am looking for a way to induce the 4.2% difference in lattice constants as a force/strain so as to see the resulting stress.

Regards,
Barry.

[HarryvL]

Barry,

Normand is right ... you really need to get FC and I will show you why

Your problem can be solved by superposition (assuming materials stay elastic). First you uniformly stretch the Si layer so it matches up with the SiGe layer (i.e. Eps = 1.4% - EDIT: oops this should be 4.2%, but the principle is the same). This induces uniform biaxial tension equal to Sigxx=Sigyy=E*Eps. You don't need FC or FEA for that . Then you "glue" the layers together and release the biaxial tension. This second stage is what we need to model and analyse with FC/CCX:

IC_1_Deformation_Max_PS.png (388.08 KiB) Viewed 1729 times

For that you create two solids and combine them with BooleanFragments (CompSolid). See my tutorial here: https://www.freecadweb.org/wiki/FEM_She ... site_Block:

IC_1_Model.png (180.12 KiB) Viewed 1731 times

With the FEM Work Bench you subsequently mesh the model:

IC_1_Mesh.png (232.25 KiB) Viewed 1731 times

and assign materials, boundary conditions and the release load I talked about earlier:

IC_1_Load.png (271.18 KiB) Viewed 1731 times

After that you are ready to run CCX and view the results:

IC_1_Deformation_Max_PS.png (388.08 KiB) Viewed 1729 times

Don't forget that you need to superimpose the biaxial tension in the Si layer to the FC results to get the final stresses. I believe this could even be done by adding them to the equation box in the Result Object (at bottom) to display the final result, but I didn't try.

FC is a fantastic tool to quickly model and analyse mechanical problems. So it's well worth the investment of time to get to know it.

Finally, here is the FC file so you have a head start:

IC_1.fcstd

(315.18 KiB) Downloaded 54 times

[HarryvL]

I see I applied 1.4% strain, whereas it should be 4.2%. Anyway, the principles and work flow are the same.

[boreilly]

Hi Harry,
Thank you very much for the detailed response. I will now download FreeCAD, learn how it works and study this exercise.
I certainly appreciate your time. I will reply again soon.

Regards,
Barry.