The Mid-InfraRed (MIR) and TeraHertz (THz) regions of the electromagnetic spectrum have a vast range of potential applications. These include security (surveillance, detection of explosives), safety (gas sensing and detection of hazardous compounds) as well as a variety of scientific and industrial applications, such as chemical analysis, thermography, medical imaging, and non-destructive testing.
The key to entering these potential applications is the ability to design optoelectronic devices specifically tailored to the particular application at hand. The design of such devices proves to be a challenging task because it involves several physical systems that have to be grasped, modeled and simulated simultaneously. Traditionally, the systems found in an optoelectronic device include charge transport, optics and heat transport. In an MIR or THz device, however, the low photon energy requires the use of low-bandgap materials and/or confined charge carrier systems; both require careful treatment, which significantly complicates the modeling and simulation task.
Due to the long wavelength of the radiation, novel types of optical guides and resonators are commonly found in the designs of MIR and THz devices, such as ring cavities, microdiscs, photonic crystals, supercrystals and micro-antennas. In our recent work, we devoloped a toolset for investigating the properties of such novel optical structures. On one hand, the analysis covers the linear solution of the optical wave equations for particular excitations. On the other hand, spectral analysis of the same equation is performed to obtain the eigenmodes of the device, which are of great interest in the context of resonating devices, typically lasers, but also detetctors where a resonating cavity "traps" the light long enough for the detector to respond.
Exact calculation of optical wave propagation would require an infinite simulation domain in most cases, which is mimicked by wrapping a finite simulation domain in a completely absorbing boundary layer, a so called Perfectly Matched Layer (PML). The PML introduces complex coefficients into the wave equation, which would otherwise be real, giving a complex-valued, non-hermitian system. This is especially important for the spectral analysis. Here, the imaginary parts of the eigenfrequencies represent "decay rates" of the corresponding eigenmodes. The vast majority of the eigenmodes are spurious modes (unphysical, caused only by the PML) and evanescent modes (i.e. modes that decay very fast). Since calculating all modes and then selecting the most stable ones is infeasible due to time and memory constraints, the calculation must be performed so that all the "desired" modes are found while calculating only a few spurious ones. The image on the right shows the simulation result of an elliptic microdisc cavity for MIR; in the wavelength vs. decay length chart the groups of stable and spurious modes can be clearly seen. Also displayed are the intensity plots of two of the stable modes.