Loss mechanisms for T-ray microwires

Analysis and design of efficient passive components for the millimeter-wave and THz bands

“…In [448], the sources of losses in dielectric subwavelength rods are breakdown concluding that losses originated by surface roughness are negligible and highlighting the relevance of bend losses. In [449], the bend problem for the same dielectric tube of [446] is afforded considering different frequencies, polarizations, core diameters and cladding thicknesses.…”

“…As an example, a very similar approach is systematically used to estimate the attenuation in dielectric waveguides, as it has been obseved in the review of the state of the art [436], [438], [442], [448], [450], [454].…”

“…At first view, one may think that at least η gives a quite good estimation of the dielectric losses, just by multiplying this factor to the bulk losses of the considered dielectric. In fact, this is a very common applied calculation [436], [438], [442], [448], [450], [454]. Note that R = 1 means that the considered mode experiences the same attenuation as a plane wave in an infinite dielectric medium of the region 1 material and this value may be firstly understood as an upper limit for R. However, in the transition between weak and confined guidance this is not longer true for enough large contrast between mediums, and some curious effect takes place.…”

“…For instance, this is the case of the coplanar waveguide (CPW) and the slotline (SL) [416]- [418], or, with less relevance, even the microstrip line [404], [409]. In a similar way, dielectric waveguides must consider this source of loss, even in the case they act confining the energy and guiding it by means of the TIR effect [21], [438], [448]- [449]. However, in both cases, radiation losses are not the main problem to look at when a design must be done.…”