Femtosecond laser writing of glass, including borosilicate, sulfide, and lead

Effectively this is a way of producing optical waveguides by using lasers to adjust optical properties of glass blocks. It uses less energy than previous methods to achieve similar or better results (speed, focus).

Key phrases:

Such a method could be used to write continuous light-guiding waveguide patterns connecting any two points within a continuous block of a suitable material, or make other optical devices, such as Bragg gratings

https://en.m.wikipedia.org/wiki/Fiber_Bragg_grating

While the mechanism of the interaction of the glass with the fs-laser is not clear, it is believed that because of the shortness of the pulse duration, the excited photo-electrons cannot thermally relax since the pulse duration is shorter than the lattice thermalization time. With high enough intensities and the inability for the electrons to relax, one can build up a relatively high electron density. It is sufficiently high to be considered a plasma. (A plasma is a collection of electrons essentially acting as a free electron gas). How the structure is permanently changed as a result of this is not known.

The invention can also be used to make interferometers or phased arrays. Also, integrated optical waveguide devices in a single glass body which utilizes multiple optical waveguide structures and paths integrated and combined together to manipulate and operate on light transmitted through the glass, such as performing a function of an inputted optical waveguide channel and an integrated optical waveguide devices which separates/combines optical waveguide channels based on wavelengths, can be made.

Some interesting photos of what's possible with that technique: http://optofab.org.au/uli.html

I've used a few FBGs from them. They were so good we had to introduce a lot of loss to avoid saturation.

In the 90s there was lots of work on storing data in 3d holograms, by using photorefractive materials. That works by interfering a reference wave and an image wave (which contains the information in bit form for example) to create 2d maps and the multiple amps through translation. You can read out the information using a reference wave only. IBM was one of the big research labs in this space.

They achieved some impressive bit densities for the time. There were even some prototypes. However, the rapid increase of drive storage densities and the alignment issues of these systems, (together with the lack of good error correction codes I suspect), meant all this was shelved eventually.

There is now quite a bit of research on using laser writing for storage (Microsoft research is working on this for example). However largely in 2d (3d is difficult because you want 2d read outs for speed and the other layers distort your image), I think they are also thinking about spinning disc type devices, so CD 2.0.

A Rotman lens might be interesting as well.