Metasurface Optics for Better Cellphone Cameras and 3-D Displays

A specially made silicon surface that can modulate light could lead to lead to a new generation of lightweight, mass-produced optical systems

2 min read
Metasurface Optics for Better Cellphone Cameras and 3-D Displays
Photo: Dr. Amir Arbabi/Faraon Lab/Caltech

Engineers at the California Institute of Technology have created a metasurface out of tiny pillars of silicon that act as waveguides for light. The way they arrange the pillars allows them to control the phase of light passing through the surface; this ability gives them control over how the light is focused, as well as its polarization, which is important for uses such as liquid crystal displays and 3-D glasses. Metasurfaces are structured planes so thin that they count as being two-dimensional; their periodic designs manipulate light in unusual ways.

“We're trying to create kind of a new platform for optics,” says Amir Arbabi, a postdoc in Andrei Faraon's Nanoscale and Quantum Optics Lab. The team described their work in the latest issue of Nature Nanotechnology.

The silicon pillars have to be somewhat shorter than the wavelength of light they're designed to manipulate. In the case of the metasurface described in the paper, the pillars are 715 nanometers tall, to handle infrared light with a wavelength of 915 nm. But they could easily be made shorter for visible light, Arbabi says. The pillars range in diameter from 65 to 455 nm, and they're elliptical in shape. The ellipses are not all oriented in the same direction; the pillars’ thickness and orientation determine how they focus and polarize the light passing through them.

Many of the same effects can be achieved with traditional optics, but that requires lining up multiple components such as lenses and prisms and beam splitters. The metasurface gets the job done with less bulk, allowing, among other things, thinner, lighter-weight cell phone camera lenses and better systems for directing the beams of industrial cutting lasers. It could also lead to novel applications. Using one of these devices, a display could switch between two polarizations and display two different holographic images. Or with an intermediate polarization, it could superimpose one image on the other. The metasruface could provide the optics for an LCD to create a 3-D display viewable from many angles without glasses.

What’s more, all of this can be done using the same lithography techniques used to build computer chips, doing away with individual fabrication and manual alignment of components. “We're trying to take these free-space components that are bulky and large and put them on a chip,” Arbabi says.

It shouldn't take much effort to move these metasurfaces from the lab to the marketplace, says Faraon. It's mainly a question of figuring out which optical system applications could benefit from the kind of mass production this technology makes possible. The array of potential applications is vast, Faraon says. “It gives you a unified framework, so you can design whatever optical component you would like.”


The Conversation (0)

The Future of Deep Learning Is Photonic

Computing with light could slash the energy needs of neural networks

10 min read

This computer rendering depicts the pattern on a photonic chip that the author and his colleagues have devised for performing neural-network calculations using light.

Alexander Sludds

Think of the many tasks to which computers are being applied that in the not-so-distant past required human intuition. Computers routinely identify objects in images, transcribe speech, translate between languages, diagnose medical conditions, play complex games, and drive cars.

The technique that has empowered these stunning developments is called deep learning, a term that refers to mathematical models known as artificial neural networks. Deep learning is a subfield of machine learning, a branch of computer science based on fitting complex models to data.

Keep Reading ↓ Show less