Discussion Forum

Hi! I'm trying to recreate the findings from this paper in COMSOL: https://doi.org/10.1016/j.mtcomm.2021.102313

In it, the researchers simulate the microwave heating of a sample material inside a WR340 waveguide. They perform a manual 1D simulation, so I'm trying to adapt it to 2D (as COMSOL doesn't have the option for 1D microwave heating simulations). For some materials, such as Alumina Cement, the simulation given by COMSOL for the temperature profile greatly resembles the one found in the paper (the temperatures themselves are a rough match). For others, the temperature ranges coincide, but not the profile. Lastly, for materials such as alpha-SiC and AlN, COMSOL gives much much lower temperatures than those found in the paper, and I can't seem to figure out why. The most important thing to achieve is a similar temperature range for every material (a similar temperature profile is not so crucial, but would be nice to get).

The entire waveguide measures 86.36x43.18x250 mm. To model it in 2D, I've created a rectangle of 250x86.36 mm, and for the input microwave I put in 800 W / 43.18 mm = 18527 W/m. I've been instructed to adjust the simulation so that the microwaves are reflected inside the waveguide, so there's no output port and there's perfect electric conductors throughout the external boundaries. As the paper doesn't model heat transfer outside the sample material (which measures 50x86.36 mm and starts at x=50 mm), there's only heat transfer in solids in the material, and thermal insulation all around it.

I think the difference in temperature may be caused by some configuration in the Electromagnetic Waves, Frequency Domain physics interface, but I'm not sure. Also worth noting is that the electric field in COMSOL roughly resembles the one in the paper, but the magnetic field is ~3 orders of magnitude smaller. I don't think this is of much importance, since I believe the contribution to heating mostly comes from the electric field. Also, I've noticed that the electromagnetic wave doesn't move when time passes. I think it should behave like the GIF in Wikipedia's article about waveguides (https://en.wikipedia.org/wiki/Waveguide#/media/File:Waveguide_x_EM_rect_TE31.gif).

If you have any suggestions, please share them, as I don't know what else to try. Also, any tips would be very much apprerciated, as I'm trying to learn all the little details of COMSOL. Thank you very much in advance!