Dear Iaroslav Filatov :

I don't think you can do this with COMSOL, since COMSOL just can use the clasical Fourier's Law and the Parabolic Heat Equation (this is the classical equation).

The parabolic Heat Equations is doesn't work for Heat Conduction for incredible short times.

My guess is that you woud have to use the elliptic or hyperbolic heat equation

But defenitively COMSOL doesn't have this models.

I don't know much about the solutions to these equations, much less about the techniques for solving them.

But I hope you can find more about this in the literature

Regards !

Thermal conductance of thin film YIG determined using Bayesian statistics C. Euler,1, ∗ P. Ho luj,1, 2 T. Langner,3 A. Kehlberger,1, 2 V. I. Vasyuchka,3 M. Kl¨aui,1, 2 and G. Jakob1, 2

Thin film YIG (Y3Fe5O12) is a prototypical material for experiments on thermally generated pure spin currents and the spin Seebeck effect. The 3ω method is an established technique to measure the cross-plane thermal conductance of thin films, but can not be used in YIG/GGG (Ga3Gd5O12) systems in its standard form. We use two-dimensional modeling of heat transport and introduce a technique based on Bayesian statistics to evaluate measurement data taken from the 3ω method. Our analysis method allows us to study materials systems that have not been accessible with the conventionally used 3ω analysis. Temperature dependent thermal conductance data of thin film YIG are of major importance for experiments in the field of spin-caloritronics. Here we show data between room temperature and 10 K for films covering a wide thickness range as well as the magnetic field effect on the thermal conductance between 10 K and 50 K.

Tavsiya qilaman!

Try to look at this:

https://www.comsol.it/model/laser-heating-of-a-silicon-wafer-13835

I had a similar challenge, which I have addressed in our recent work: Article Ultrafast Photoinduced Heat Generation by Plasmonic HfN Nanoparticles

Indeed, I've used Lumerical to calculate the absorbed power profile and COMSOL to calculate the heat transfer dynamics using a two-temperature model. I've used the same methodology for ultrafast heating of gold arrays and nanoparticle arrays, and it should work for your problem too.

For modelling the physics, you must setup a two-temperature model (for the coupled electron and the lattice temperatures) under your thin film boundary conditions that is properly considering the temperatures and heat-flow at the interfaces. An example was pulished here

Article Partial ablation of Ti/Al nano-layer thin film by single fem...

You can implement you model either in COMSOL or Lumerical.

It considers a 1D-model that should work in your case if your beam radius is much bigger than the energy depostion depth (which is very likely the case). If so, you can "construct" your radial fluence variation by simulations with different (constant) fluence values.