AbstractSilicon photonics has been established as a mature and promising technology for optoelectronic integrated circuits, mostly based on the silicon-on-insulator (SOI) waveguide platform. However, not all optical functionalities can be satisfactorily achieved merely based on silicon, in general, and on the SOI platform, in particular. Long-known shortcomings of silicon-based integrated photonics are optical absorption (in the telecommunication wavelengths) and feasibility of

more »

... cted lasers (at least at room temperature). More recently, high two-photon and free-carrier absorptions required at high optical intensities for third-order optical nonlinear effects, inherent lack of second-order optical nonlinearity, low extinction ratio of modulators based on the free-carrier plasma effect, and the loss of the buried oxide layer of the SOI waveguides at mid-infrared wavelengths have been recognized as other shortcomings. Accordingly, several novel waveguide platforms have been developing to address these shortcomings of the SOI platform. Most of these emerging platforms are based on heterogeneous integration of other material systems on silicon substrates, and in some cases silicon is integrated on other substrates. Germanium and its binary alloys with silicon, III–V compound semiconductors, silicon nitride, tantalum pentoxide and other high-index dielectric or glass materials, as well as lithium niobate are some of the materials heterogeneously integrated on silicon substrates. The materials are typically integrated by a variety of epitaxial growth, bonding, ion implantation and slicing, etch back, spin-on-glass or other techniques. These wide range of efforts are reviewed here holistically to stress that there is no pure silicon or even group IV photonics