← Back to all notes
R

Roel Huis in ‘t Veld

Director, Kriya Materials · Week of November 16, 2026

The combiner is moving into the lens, and the coating has to move with it

Karl Guttag published the third part of his 2026 AR year in review on 13 November, covering TDK's meta-optic mirror, the Fourier Optics metasurface and a few more radical combiners. My takeaway as a coatings supplier is simple. The combiner is moving inside the ophthalmic lens, and whoever embeds it will need index-matched materials, encapsulation and anti-reflection on a part that behaves nothing like a flat waveguide. The physics of these combiners is clever. The physics of the interfaces around them is where the image quality gets lost. Take TDK first. Guttag describes the Meta-Optic Mirror as a nanostructure of 200 nanometres to a micron, made in a semiconductor-like process on a thin substrate and then embedded in an ordinary eyeglass lens. It redirects a laser from the temple into the eye for direct retinal projection, green only for now. Fourier Optics looks similar from the outside but is built differently: planar layers stacked, sliced diagonally with a diamond saw, then embedded in a lens. Fourier claims 96 percent transparency, a 20 degree field of view, and about 3 dollars for the metasurface plus 5 dollars to put it in the lens. Now the part the big outlets skip. Each of these elements sits inside a polymer lens, which means new buried interfaces: lens half, fill or adhesive, substrate, nanostructure, and back again. An index step of 0.05 between two layers reflects about 0.03 percent at normal incidence, but the laser arrives at a grazing angle and the wearer looks through the stack at a bright sky. What the eye picks up is not the reflection but the outline of the embedded element, a faint rectangle in the lens. Guttag notes that Fourier's own through-the-optics image shows scattered light, and that early prototypes typically lack the coatings and precision to show what the technology can do. That is the work. Then there is temperature. Polymers shift their refractive index by roughly one part in ten thousand per degree, glass and inorganic nanostructures by an order of magnitude less. A lens matched perfectly at 20 degrees in the lab is 0.004 off on a dashboard in July. A waveguide can live with that. A combiner buried in a plastic lens with laser light grazing it cannot. What customers will ask for next is not a lower index as such but a cured material whose index sits at a specified value, plus or minus 0.005, from minus 20 to plus 80 degrees, with no yellowing and no haze, bonding to a hard inorganic substrate on one side and CR-39 or polycarbonate on the other, with an anti-reflective stack on top. That is a formulation question, not a nanophotonics question, and a small coating house answers it faster than a chip fab. Guttag closes by saying that when the rational ideas do not work, you start on the irrational. I would put it differently. The irrational idea is to spend a year on a 3 dollar metasurface and a week on the 5 dollar lens around it. Sources KGOnTech (Karl Guttag), “AR in 2026: Year In Review (Part 3: Novel Combiners and Coherent LBS)”, 13 November 2026. https://kguttag.com/2026/11/13/ar-in-2026-year-in-review-part-3-novel-combiners-and-coherent-lbs/