Optical Wavelength Cloaking Carpets

Michal Lipson and pals at Cornell University and Xiang Zhang and buddies at UC Berkeley say they have both built cloaks that are essentially mirrors with a tiny bump in which an object can hide. The cloaking occurs because the mirrors look entirely flat. The bump is hidden by a pattern of tiny silicon nanopillars on the mirror surface that steers reflected light in a way that makes any bump look flat. So anything can be hidden beneath the bump without an observer realising it is there, like hiding a small object under a thick carpet.

Cloaking at Optical Frequencies : Cornell University

Cloaking principle of the fabricated device. A planar mirror forms an image equal to the object reflected (a), but when the mirror is deformed, the image is distorted (b), allowing an external observer to identify the deformation. Our cloaking device – shaded area in front of the mirror – corrects the distortion in the image, so that the observer no longer identifies the deformation in the mirror, nor an object hidden behind the deformation (c).

The cloak operates at a wide bandwidth and conceals a deformation on a flat reflecting surface, under which an object can be hidden. The device is composed of nanometer size silicon structures with spatially varying densities across the cloak. The density variation is defined using transformation optics to define the effective index distribution of the cloak.

These results represent the first experimental demonstration of an invisibility cloaking device at optical frequencies. The bandwidth and wavelength of operation of the device is limited by the bandwidth of operation of the distributed Bragg reflector. This bandwidth is large, 950 nm, around a wavelength of 1500 nm due to the large index contrast between silicon and SiO2. Such a cloak could in principle be reproduced over much larger domains, using techniques such as nanoimprinting, for example, enabling a wide variety of applications in defense, communications, and other industries. Note that in this paper we show how the trajectory of light can be manipulated around a region to render it invisible. Using transformative optics in a similar fashion to the one used in this paper, one could do the opposite – concentrate light in an area. This could be used for example for efficiently collecting sunlight in solar energy applications [concentrate light].

Dielectric Optical Cloak: UC Berkeley

the first experimental realization of a dielectric optical cloak. The cloak is designed using quasi-conformal mapping to conceal an object that is placed under a curved reflecting surface which imitates the reflection of a flat surface. Our cloak consists only of isotropic dielectric materials which enables broadband and low-loss invisibility at a wavelength range of 1400-1800 nm.

The experimental demonstration of cloaking at optical frequencies suggests invisibility devices are indeed within reach. The all-dielectric design is isotropic and non-resonance based Wavelength dependence of the carpet cloak. Plotted is the intensity along the output grating for a curved reflecting surface (A) with cloak and (B) without the cloak. The cloak demonstrates broadband performance at 1400 nm – 1800 nm wavelengths. Distinct splitting of the incident beam is observed from the uncloaked curved surface due to the strong scattering of the original beam. Therefore promising a new class of broadband and low-loss optical cloaks. It should be noted, that this methodology can also be extended into an air background by incorporating non-resonant metallic elements to achieve indices smaller than one. Furthermore, the quasi-conformal mapping design and fabrication methodology presented here may open new realms of transformation optics beyond cloaking.

About The Author

Add comment

E-mail is already registered on the site. Please use the Login form or enter another.

You entered an incorrect username or password

Sorry, you must be logged in to post a comment.


by Newest
by Best by Newest by Oldest


Is there any better source that describes with more accuracy this technology? The claiming that they use optical litography in 12nm is nearly unbelievable. I'd like to see something more substantive.


15 years? 12nm is at most 5 doublings, and by "2 years per doubling", that's just 10 years. Two years per doubling is on the conservative side, but it's getting harder to shrink feature size.

We're pretty firmly at 45nm already, so 12nm is about 4 more generations (32, 22, 16, 12) or 8 years. We're nearly a year into 45nm CPUs, so maybe 2015 for introduction of 12nm CPUs?

So we might expect 8x to 16x more processing power, at least in the form of more processing cores.

Except heat isn't falling with transistor area any more. Maybe we'll get 4x more processing at the same power usage?

So maybe laptops will get 4x faster, typical desktops 8x faster, and super-high-end-damn-the-heat-full-speed-ahead systems maybe 16x faster (probably only for the GPU, as CPUs are focusing strongly on power efficiency).

Cooled docking stations with built-in high-end graphics will probably let lap-tops get to about half the highest performance systems...