Another Meta

Another Meta
​

ederico Capasso, a former Bell Labs (NOK) applied physicist who was one of the inventors of the quantum cascade laser, and now a Senior Research Fellow at Harvard, came up with the idea of using metasurfaces for flat optics in 2016 and has been leading research in this new field since, with over 70 US patents and over 500 peer-reviewed journals (not all on metasurfaces).  The idea of metasurfaces, or sub-wavelength thickness structures that are placed in a horizontal manner, is similar to that of more typical optical lenses that focus light using refraction, while metasurface optics use small nanostructures to scatter light and by capturing and changing a number of optical characteristics, including phase, polarization, and spectrum, can be used for a number of optical processes that are typically done with glass or similar medium.  The shape and patterning of these nanoobjects determines their optical characteristics, opening an almost infinite range of possible shapes, sizes, and materials for these lenses, which are typically in the range of 100nm thick, or 1/100,000 of the equivalent optical lens.

Opticqal Lens Refraction Focus - Source: Nature - Optical metasurfaces: new generation building blocks for multi-functional optics, Neshev, Dragomir, Aharonovich, Igor

Metasurface Phenomena with control of color and polarization - Source: See above.

What makes metalenses important is that they are produced using photolithography or ALD (Atomic Layer Deposition), both of which are used in the manufacturer of semiconductors and a few years ago MIT scientists developed a computational technique that determined the optimal makeup and arrangement of metalens elements, enabling designers to meet specific optical parameters, reducing prototype and rework timelines.  As, in theory, the process by which the metrastructures scatter and re-emit the source light is !95% efficient, such lenses are thought to have an advantage over conventional glass lenses in terms of cost, size, weight, and performance.  Further a single refractive lens (glass) can project an undistorted image onto a curved surface, but in order to project on a flat surface, additional lenses are needed to eliminate the curvature distortion.  As much in the image sensor world is flat, only a single metalens is needed for a flat (2D) surface, such as in a smartphone, eliminating the need for the lens ‘bumps’ seen on most smartphones.
Companies like Imagia (pvt), NIL Technologies (pvt), and Metalenz (pvt) are developing applications that try to simplify the optical stacks used in a number of consumer electronics’ applications.  STMicro (STM) is using a metalens in its 6/22 release of a ToF (Time-of-flight) module, using Metalenz technology that is produced on STM’s silicon lines.   The device uses two metalenses that substitute for the more typical optical lenses and has a power consumption of 31.3% less than that of the optical version in continuous mode.

3Metalenses Structures - Titanium Dioxide nanofins that operate at 660 nm (red) & Silicon nanobeams that create spot focal point – Source: Science 352, 1190-1194 (2016) & 345, 298-302 (2014)

​While metalenses sound game-changing, and they are to a degree, they have two flaws.  The first is that they are small.  Achromatic lenses, meaning that the color remains the same through all parts of the lens, is relatively easy to do with glass, essentially making the lens thicker as it gets larger.  Metalenses are achromatic when small (few hundred microns), but have color aberrations as they get larger, which limits them to very small applications.  There have been positive lab results using a large number of such lenses to stich together a wide-angle image, but with no commercial application yet.  The efficiency of metalenses is the 2nd flaw, and while such is over 80%, meaning 80% of the light that enters the metalens is ’converted’, Fresnel lenses (plastic on glass) and pure glass lenses are higher, and as noted below, this can be an important factor in some applications.
One area of particular interest is VR, where complex optics add considerable weight and bulk to headsets.  Flat metalenses can be less than 1mm thick and are rectangular, which matches the format of most digital image sensors.  With pixel pitch below 10um, such lenses allow pixel or even sub-pixel beam steering and can replace optics that require a number of lenses or steering mechanisms.  Since VR displays are small, typically 1” or less, the size issue is less of a factor, but efficiency is more so, as the clarity of the image in VR is key, but considerable research continues and as better materials and patterns are found, there is considerable hope that higher efficiencies can become available, with a recent lab process having 94% efficiency, albeit not in a commercial setting.  The use of DUV (Deep Ultraviolet) patterning has led to higher efficiencies but work still needs to be done to move those results from the lab to commercialization.
All in, if metalens research is able to conquer a few limitations, they open a whole new world for AR/VR and a host of other applications that rely on physical optics.  Glass manufacturers and optical component companies do not have to worry about their business yet, as metalenses are still primarily a highly specialized field that has potential but no guarantee that it can be competitive with 3-dimensional optics, but every ounce of weight that can be removed from a VR headset will make them more compatible with the general public and metalenses are trying to fill that bill.