Learning Forum

[Felix Lichtenegger]

Hello everybody.
In this nice tutorial a method for simulating a metalens is presented, the so called near-field stitching:
https://optics.ansys.com/hc/en-us/articles/360042097313-Small-Scale-Metalens-Field-Propagation 

Instead of simulating the whole metalens, the  nearfields from different unit-cell simulations is stitched together and then propagated to the farfield. In the tutorial and in my own sucessful simulations, perpendicular incidence of the plane wave is used for every unit-cell. 

If I want to simulate the metalens for, lets say, 10° incidence of a monochromatic plane wave, I assume I have to simulate each unit cell with 10° incidence and use bloch boundary conditions. Then again, the near field is saved, stitched according to the phase profile of the lens and propagated to the farfield. Is this assumption correct or are there other effects to consider?  

First tests show an absence of spatial shift of focal points when using the near-field stitching approach, which brings me here. A full lens simulation without stitching, shows a spatial shift of focal points as one would expect.  

Thanks in advance for any help.

[Guilin Sun]

This is a very interesting discovery! 

I did not find the reference paper for stiching method. But I guess it may not conisder the off-axis effect: it may assume that the propagation has the same axis as the incident or the desired diffraciton order. 

I guess this is a theoretical problem. You have verified with direct simulation so such stiching method might need to be improved.