Transformation Optics Lens

[katto01]

Hello! I recently found Optiland, and I am very excited to begin learning it. It looks great! I am working my way through the tutorials. Before I spend too much time, I wanted to ask a few questions:

I am tying to apply to to generate a flat lens that performs similarly to a Luneburg Lens. The implementation could be first in 2D where an array of rectangles would be assigned a certain dielectric constant. A plane wave would be incident and the maximum files location would be identified. Once the focal points (for a few plane waves with various incident angles) are identified, an excitation would be posititoned in these places, and the directions of the outbopund rays would be calculated. Is this possible to do this 2D and then 3D analysis using Optiland?

Thank you

[HarrisonKramer]

This is not really what Optiland is built for right now. It is a sequential ray-tracing tool: rays hit a defined series of surfaces in order, rather than propagating through an arbitrary spatially-varying dielectric medium like your rectangle-array idea would need. The closest related work in progress is GRIN (gradient-index) material support (#337), which handles continuous n(x,y,z) profiles within a surface, but that is still not the cell/metamaterial-array model you are describing. If you want to prototype the field-tracing/focal-point-identification part externally and then import the resulting ray directions as a phase/deflection profile on a surface, that might be a workable hybrid, but full transformation-optics synthesis is not on the roadmap currently.