What is a grating coupler?
Grating couplers are periodic lines that are etched into
the top of the waveguide, which re‐direct the light from
the optical fibre into the waveguide.
What is an edge coupler?
An edge coupler is a device that couples light from fibre to the chip using inplane/butt/ edge coupling.
Tapering is a vital component for these devices.
Advantages of edge coupling
Edge coupling, in essence, is the inverse taper method.
Advantages of grating couplers
Disadvantages of general edge couplers
Relatively incompatible with wafer-level testing
Requires high-quality facets on the chip
Disadvantages of Grating Couplers
Chip Facet Considerations in edge coupling
Chip facet: the chip facet must be made very smooth, to reduce unwanted scattering. This can be done by polishing, cleaving, or dicing…all of which can be expensive processes.
Describe Regular Planar Taper

Describe 3D Taper

Describe 3D Taper Etch Steps

Edge Coupler Consideration
Even if you match the mode shape of the waveguide to the mode shape of the fibre well, if the waveguide effective index is greater than that of the silica glass that the fibre is made of (n=1.45), there will be a reflection at the interface.

Give Equation for Fresnel Reflection

What is an anti-reflection coating?
Thin-film coating, which has a refractive index between that of two materials.
Used to reduce the reflection at the interface of the two optical materials.
How does Rayleigh’s film anti-reflection coating reduce reflection?
To minimise the reflection, the refractive index of the
intermediate layer should be the geometric mean of the
refractive indices of the two outer layers:
Reflection can be further suppressed by including more
layers between waveguides of different widths.

Describe interference Coatings
which also involves inserting a thin film at the interface.
There will be a reflection from the first and second boundaries of
that thin film, which will then interfere with each other. The
thickness of the thin film can be chosen such that they interfere
destructively, minimising the reflections.
For this there should be a 𝜋 phase shift (Δ𝜃) between the two
reflected beams:
Δ𝜃 = (2πn1L)/𝜆
Hence π = (2πn1L)/𝜆
Rearrange, L = 𝜆/2n1
For Normal Incidence, Layer thickness
t = L/2
Therefore t =𝜆/4n1
