Discussion Forum

How does annealing affect the stress of a Si wafer? Is there an oxide layer formed? There needs to be some sort of structural change otherwise the wafer will elastically return to its original state when cooled.

sorry,it's my fault
actually the structure of the wafer is Au-alumina-Si
the thickness of the Au film is 50-100nm,
and I need to analyse the internal stress of the Au film after annealing

thanks for your reply

So the crystalline structure changes at temperature, for example grain growth in the gold resulting in tensile strain, then when it's cooled there's an additional stain from thermal mismatch? Or maybe there's viscoelastic flow in the Au layer which results in stress reduction at the anneal temperature which is then adjusted by thermal expansion mismatch? If so there I can't help you because I'm completely ignorant of any state change modeling in Comsol (I'm just learning how to use it myself). But maybe the support guys can help. I think the simplest approach would be to somehow set the stress state of the film (which would be biaxial if it's a planar film) at the critical temperature then reduce the temperature from there with the material then treated as a solid.

appreciate for your reply anyway

Hi

there should be a Model library example of thermal stress from a bi-domain structure, and how to continue a study from the initial stress build-up. If I remember right define the geometry at the "hot" temperature, set the stress free Temperature to the "hot temperature", let cool down. then start a new study from the deformed/stressed cooled down state.

If you want to structure one layer, draw the structures and then turn the Young modulus down to "zero" via a variable and you will get the results, at best for a structured, pre-stressed system

--
Good luck
Ivar

thanks

should the initial temperature be the annealing temperature?or the atmospheric temperature?