New to COMSOL - Issues with Modelling a MEMS Coated Micromirror (2D)

Hi,

I am new to COMSOL software, however I have started building a 2D model of a MEMS Micromirror with a thin film coating of SiO2. My aim is to match experimental results regarding the radius of curvature given a change in temperature. To do this I will have to fiddle about with Young's Modulus and CTE, since these are not well known for thin films.

The file is attached, and is able to run error-free. However I am very sceptical of the results produced. I have a 2-step process, the first being application of the coating, and then the second just a simple step up or down in temperature. The results show a curvature in the upward direction for the first step, then a very large change in curvature downwards, even though the temperature change should be smaller and in the same direction (Ttop is deposition temperature, Troom is cooling to room temperature, Tchange is change in room temperature). So I am questioning my current setup, or perhaps I am misunderstanding some of the features.

Things I am unsure of:

- Strain Reference temperature for the second step
- Structure of the solver
- Initial values for second step

Any advice people could give me regarding this would be very much appreciated. If you need any more details I will provide them (if I know how!).

Thanks,
Alan