Learning Forum

[anindya]

Dear Experts,

I have been doing the heat simulation of a certain material layer (material x) on silicon dioxide substrate and the structure is pity simple.Material X is in between the two Ag electrode layer and the total arrangement is on the substrate.I wanted to see the increase in temperate due to the effect of applied voltage inside the material layer x only in order to test the crystallization temperature of that material x.I have build the material in the database by using its thermal datas (such as conductivity , specific heat etc) available in the literature.Even doing everything as per the protocol such as simulation region, Voltage probing, boundary conditions, temperature monitors, I am not getting the accurate values of crystallization temperature that is desired.For my case the crystallization change can be monitored in terms of refractive index variation which can be seen by loading the heat solver data into the Lumerical FDE solver.By lietature the crystallization change should occur at a temperature of around 450K by I am getting it at 365K.Can you please suggest what kind of necessary corrections should be done here? I will look forward to your expert opinion.

Thanks and regards,

Anindya Bose

[anindya]

Dear Mr Gullin Sun,

Can you please address my issue.I feel that I am not getting the accurate result in Heat Solver because of not tuning the accuracy parameters. Can you please suggest some accuracy parameters which can be carefully tuned to get an absolute thermal data of the resulting device.

 

Thanks and regards,

Anindya Bose

[Guilin Sun]

Sorry to reply you late.

crystallization temperature is for bulk material, I believe. When it is nano scale, the thermal properties are different from bulk. I read a paper saying that the value can change to be 2/3. Thus using the bulk properties to predict the nano-scale structure is not accuate.

I would suggest you to check:
if the HEAT settings are proper, including material properties;

You mentioned FDE. that means you are simulating a waveguide and find the neff change due to temperature.  make sure the optical material property is proper.

If you know the dn/dT, probaly you can manually calculate the refractive index change without using the waveguide mode. This thermal coefficient also changes with thickness. 

As long as the simulation settings are correct, the result is reliable, provided that in experiment the meterial properties are the same as what you simulate. The HEAT solver is linear so it is unlikely the result is wrong when the settings are correct. eg, the boundary conditions are correct. Please note that the HEAT solver region is quite small in general so make sure the temperature is correct.