Hi,
I was trying to implement frequency dependent complex dielectric permittivity in COMSOL. The simulator considers permittivity in the form e(w)=e'-je"(w). The complex part is supposed imply the lossy term or joule heating, right?
I have been using the frequency dependent permittivity from Otter's and Johnson & Christy in e(w)=e'+je"(w) form. But the imaginary part from these references are supposed to be used as positive or negative? the COMSOL Blog says, you should consider the imaginary part as negative, but when I do that the results seem wrong. Could someone please help me with this?
sign convention for imaginary part of both permittivity and refraction index follows the sign convention of forward propagating wave. if exp(-jkz) is used, then the imaginary part has negative sign. In Comsol --->wave equation, Electric you can see it uses negative sign convention for both refraction index and forward propagating wave in the z direction.
So, it means that if I am using exp(-jkz) wave propagation and since COMSOL uses e(w)=e'-je"(w) form of dielectric permittivity, I should input positive value for imaginary part e", right?
May I ask, the complex dielectric permittivity of metals and even other elements like dielectrics that we get from literature and previous reports, do they follow a specific form e(w)=e'-je"(w) or e(w)=e'+je"(w) by default, or does it vary from book to book?
I have seen both of the forms (for example for graphene) and as I said it depends on the sign convention of the forward propagating wave.
One should always do a Wikipedia search nowadays as it has gotten very good. See https://en.wikipedia.org/wiki/Permittivity for an explanation of the different conventions..
- Badges
- Science topic