consult for dielectric function fit

I am confused by one question, which is very bad dielectric function fit results when I set the imaginary part is zero. How to attain better fit?

Can you upload the file. I can fit the data and then instruct you on how to achieve the same result.

Hi, I fit the data by imaginary weight,but I am confused that what’s meaning of imaginary part? And there is still imaginary data when I set it to zero.


the imaginary part of refractive index is associated with loss.
When we fit, we use physical model as all the materials have loss in some wavelength range according to KK relationship.
if in the desired wavelength range the imaginary part is essentially zero, you can fit them with loss on the order of -5 or smaller so the result will not be affected however the material model is physical.

1 Like

Hi,I always failed to fit material diff_img_0_al in my file. Now, I submit file to you, please help me check the reason.
Thanks again

best wishes

al_110_150_280_0_img0.fsp (290 KB)

It seems that your material is not a natural material. Could you please verify how did you get the data and if it is a natural material? The MCM is the best for natural material.

I suggest that you can try to set the imaginary part to a very small value like 1e-6 rather than zero.

Hi, yes.It’s not a natural material. I just set Al dielectric function imaginary part to zero.
It’s a try to understand a model. It doesn’t matter that fitting the curve, successfully.

Thanks for your kindly reply.

In this case, it can be difficult to get a good fitting, in particular you omitted the imaginary part and set it to be zero.maybe you can try:
1: use the original data (not leave the imaginary part to be zero);
2: sweep wavelength by fixing the mesh generation. Since the data is not fitted, the result may not be smooth.

1 Like