Several strategies have been proposed to circumvent this constraint and engineer the optimum refractive index for a specific application. The classical approach is to play with material composition. A well-known example is silicon oxynitride (SiOxN1-x) glass, in which the amount of oxygen and nitride can be balanced to ideally tune the refractive index continuously from that of silica (nSi02 ~1.45) to that of silicon nitride (nSi3N4 ~1.98). An alternative strategy, recently demonstrated in Si02/Si3N4 TriPlex technology, makes use of multilayered waveguides, where an “artificial” index contrast is tailored by changing the cross-sectional geometry only, without modifying material composition.
In the work by Cheben et al., a third powerful method is proposed, in which neither the material nor the cross section of the waveguide is modified. The waveguide is longitudinally patterned with a subwavelength grating (SWG), consisting of segments of a high-refractive-index core material interlaced with a lower-refractive-index cladding material. Since the refractive-index contrast can be changed by simply controlling the grating period, SWG waveguides with different optical parameters (mode confinement, effective index, chromatic dispersion, and so on) can be realized on the same chip. This approach nicely fits the fabrication processes of planar lightwave circuits and represents a noteworthy step forward with respect to the above-mentioned methods, which are not suitable for controlling the refractive index in a specific location of the chip.
At first sight, one could argue that the concept of segmented waveguides is not original in photonics, as it was proposed approximately 20 years ago for the realization of 2D mode expanders for low-loss coupling with optical fibers. Yet, Cheben’s group has borrowed the idea to find a solution to a more general problem: that of circumventing the tight limitation set by the fixed refractive index of a photonic platform. By optimizing the periodicity of the grating, they realized silicon SWG waveguides with a dispersion characteristic (around a wavelength of 1550 nm) matching that one of unpatterned strip waveguides with an equivalent core index of 2.65 and 2.03. They claim to be able to change it at will in principle, covering the entire range from silicon (nSi ~3.5) to the SU8 polymer cladding material (nSU8 ~1.6)
Besides the ability of synthesizing arbitrary refractive indices, what is really impressive in their findings is that, apparently, no great price is paid in terms of loss. The realized waveguides exhibit a propagation loss as low as 2.1 dB/cm, not too far from the value of 1 dB/cm of state-of-the-art deep-etched silicon waveguides. Although being consistent with Bloch theory of propagation in periodic structures, the light capability of travelling almost undisturbed across so many strong discontinuities is fascinating. From a more practical point of view, it means that SWG waveguides are ready for applications and competitive with conventional waveguides.
What we learn from this work is a new strategy to create versatile building blocks for the realization of advanced photonic integrated circuits. What we can do now is to put the right waveguide at the right place within our optical chip. This was a missing degree of freedom in photonics, but now it seems possible.
You must log in to add comments.