Hi all,
I’m having trouble with the idea of a complex refractive index (CRI) defined for a dielectric_dielectric interface.
I have two volumes V1 and V2 in my simulation with a surface(S1) defined for the interface of V1 and V2. V1 and V2 are both dielectric materials (V1 is actually liquid argon and V2 is PMMA acrylic). V1 contains the incident and reflected photons and V2 contains the transmitted photon. Now V2 has a CRI defined as one of its optical properties. If I’m not wrong, the CRI being defined for a volume doesn’t affect the reflection/absorption/transmission of the photon at the interface. Please correct me if I’m wrong here.
As for the surface S2 defined for the interface - it is defined to be a dielectric_dielectric type surface (with the “unified” model and a “ground” finish). Now, S2 is made of a material (PMMA acrylic) that has the CRI index defined as well. That’s where my confusion is arising.
Does it make sense to have a CRI defined for a dielectric_dielectric interface? (referring to Fowles’ optics book, CRI is usually used for dielectric_metal interfaces. I also referred to this article on arxiv - https://arxiv.org/pdf/1612.05162.pdf. Refer to page 8 of the article. Although I should mention that I’m not completely sure about the validity of the statements in this article. Hence asking the community here.)
The followup question to the above is this. Let’s suppose that defining a CRI for a dielectric_dielectric interface does make sense (if the second volume is absorbing, it may be fine to define a CRI for that interface). In that scenario, when I traversed through the G4OpBoundaryProcess.cc code (http://www.apc.univ-paris7.fr/~franco/g4doxy/html/G4OpBoundaryProcess_8cc-source.html), I found that it calculates the reflectivity for the surface (Line 358). Calculating the reflectivity makes sense. But when I scroll down a bit more, I find that for a dielectric_dielectric surface, the code is cooked up such that there could be absorption with the probability of (1-reflectivity) (Line462). This doesn’t make sense to me. From a purely physics viewpoint, how does it make sense to have absorption of the photon at the surface itself? From what I’ve read, GEANT4 has 3 possibilities for a photon at any interface - reflection, transmission, and absorption. I don’t see how the absorption of a photon can be considered a boundary process. (The photon/electric field can transmit and then be killed in the second volume, but absorption AT the interface doesn’t make sense to me)
Even if it is argued that it is an effective model and that having absorption for an interface with a CRI makes sense, there is a similar problem with the user-defined REFLECTIVITY parameter. If, for a dielectric_dielectric interface, the REFLECTIVITY parameter is defined, it behaves the same way as the calculated reflectivity in the CRI case, i.e., it implies that the photons at the boundary are absorbed with a probability of 1-REFLECTIVITY (L461 of the G4OpBoundaryProcess code). How does the absorption in that case make sense?
Any light shed on these issues will be much appreciated. Thanks.