If you think metamaterial has only “invisible
cloak” and/or “negative index of refraction” in her hat, think again. Researchers from the school of
engineering and applied science at Harvard utilize a powerful feature of the metamaterial
to create ultra-thin and flat lens that is diffraction limited. They also
create a flat axicon as another example when pioneering in this field (A quick
glimpse on the Axicon: A lens with a flat surface on one
side, and a conical shape on the other, has the ability to focus a Gaussian beam
into a Bessel beam at the focal region, and create hollow ring beam shape in
the far field).
To focus light, we need to create a
converging spherical wavefront, or at least, a wavefront that is converging. To
do so, we need to introduce different phase retardation on different portion of
the incident light. A spherical lens does so by letting the light pass through different
amount of material. For example, when light is passing through the center of
the lens, it lags behind compared with that passes through the edge of the
lens. As a result, there is a phase difference between them. This phase difference,
or phase retardation between them, produces a converging wavefront. However, when
a plane wave like light passes through a spherical lens, it suffers from spherical aberration. That is to say the spherical lens
does not produce a perfectly converging spherical wavefront which is required
for the light to focus tightly. The light exiting from the edge of the lens
suffers stronger deviation. The best way to solve this so far is to use an
aspherical lens to correct this imperfection.
A puzzle like this can be solved
beautifully by the use of the metamaterial. If we are able to create
metameterials which can introduce different phase retardation, arrange them in
a way such that they produce a converging spherical wavefront, we can have a
flat, thin and aberration-free lens. This idea is realized and
performed nicely finally. Researchers create gold antennas with different
shapes. When impinged by the light, each shape is able to create different
amount of phase retardation while the scattering intensity remains similar. From
figure 1, we can see clearly that only the phase of the impinging light is
shifted while the scattering amplitude remains close to each other. The lens is
then realized by patterning silicon wafer with these gold nano-antennas using
electron beam lithography. The antennas with different shapes are arranged in a
ring pattern which would give the desired phase retardation. The spacing between
the antenna arrays is covered with silver and titanium which completely reflects
the fraction of the incident beam that is not impinging on the antennas. The
wavelength of the incident light is 1.55 um, and the diameter of the lens is
just merely 0.45 mm.
 |
|
Figure 1. (a) The small antennas at work:
different shapes of gold antennas are able to produce phase retardations. The
sizes of the antennas are ranging from 85-180 nm with the width of 50 nm. (b)
The experimental layout. (c) The arrangements of the antennas on a lens made of
silicon wafer. The antennas are spaced ~ 500 nm to each other. Courtesy of F. Aieta,
P. Genevet, M. Kats, N. Yu, R. Blanchard, Z. Gaburro, and F. Capasso in NANO
letters.
|
The lens has one more special
feature, it takes linear polarized light and focuses the light in a cross
polarization way. In other words, if you put in x-polarized light, the resultant
converging wavefront is y-polarized, and vice versa. In addition, to create a
converging spherical wavefront, the phase retardation has to be a hyperbolic
function vs. the radius of the lens. And to create an axicon, the phase
retardation has to be conical. A contour plot of the phase retardation vs. the
radius of the lens is shown in figure 2. The results compared to the calculation
are very promising, as shown in figure 2, lower part.
 |
 |
By arranging the antennas differently, we can create virtually all kinds of lens. For example, if the array of the antennas has translational symmetry, we can have an elliptical lens. By tailoring the phase front to have great curvature, we can have a high NA objective. In addition, they are flat, thin and easy to assemble. There is a caveat to be aware of. At the moment, only 1% of light is transmitted, since the antennas only cover small amount of the area on the wafer. In order for the real application, the lens efficiency has to be decent, maybe around 80%. The researchers are trying diligently to increase the efficiency, presumably by increasing the density of the antennas.
There are quite a few other interesting flat lens designs using nanotechnology. A few examples are nano-holes and nano-slit lens. For the reflective optics, flat dielectric grating reflector with nano-scale patterns is also attractive. Feel excited? Just click on the links.
DISCLAIMER
DISCLAIMER
The opinions expressed herein are those of the author and do not represent the Optical Society of America (OSA) or any OSA affiliate.
