1
SCIENTIFIC
RepoRts
| 6:32803 | DOI: 10.1038/srep32803
www.nature.com/scientificreports
Multiwavelength metasurfaces
through spatial multiplexing
Ehsan
Arbabi
, Amir
Arbabi
, Seyedeh
Mahsa
Kam
ali, Yu
Horie
& Andrei
Faraon
Metasurfaces are two-dimensional arrangements of optical scatterers rationally arranged to control
optical wavefronts. Despite the significant advances made in wavefront engineering through
metasurfaces, most of these devices are designed for and operate at a single wavelength. Here we show
that spatial multiplexing schemes can be applied to increase the number of operation wavelengths.
We use a high contrast dielectric transmittarray platform with amorphous silicon nano-posts to
demonstrate polarization insensitive metasurface lenses with a numerical aperture of 0.46, that focus
light at 915 and 1550
nm to the same focal distance. We investigate two different methods, one based
on large scale segmentation and one on meta-atom interleaving, and compare their performances.
An important feature of this method is its simple generalization to adding more wavelengths or new
functionalities to a device. Therefore, it provides a relatively straightforward method for achieving
multi-functional and multiwavelength metasurface devices.
Metasurfaces have been of great interest in recent years as they provide optical wavefront control with high spa-
tial resolution, in a thin and flat form factor
1–5
. High efficiency, capability to control polarization and phase, and
compatibility with the standard nanofabrication techniques have made dielectric transmittarray metasurfaces
very attractive for various applications in different parts of the optical spectrum
3,6–22
. Conventional optical devices
such as blazed gratings, lenses, and orbital angular momentum generators
6–8,10,11,13–16
, simultaneous polarization
and phase controllers
12,23
, and flexible and tunable devices
18,24,19
have been realized with high efficiencies using
dielectric transmittarrays.
Similar to other types of diffractive optical devices, metasurfaces which deflect light suffer from large
chromatic dispersion
25–28
. Therefore, these types of metasurfaces are usually designed and operate at a single
wavelength. Recently, multiple methods have been proposed for realization of multiwavelength metasurface
devices
27–32
. Most of these devices are polarization sensitive and suffer from low efficiencies
27,29–32
.
Spatial multiplexing has been used for enhancing the number of operation wavelengths
33,34
or adding new
functionalities to optical devices
35
. Here we show multiwavelength metasurface lenses based on spatial multi-
plexing with two different approaches: large scale aperture division and meta-atom interleaving. We use dielectric
transmittarrays based on amorphous silicon (
α
-Si) to implement the two different methods, and experimentally
demonstrate lenses that focus light with wavelengths of 915
nm and 1550
nm to the same distance. The methods
introduced here can readily be generalized to more than two wavelengths, or to devices that perform different
functions at different wavelengths, or at the same wavelength.
Concept
Two metasurface lenses designed for two different wavelengths (Fig. 1a,b) can be combined through dividing
the metasurface aperture into macroscopic areas, resulting in a multi-sector device as shown in Fig. 1c. A less
obvious method is to interleave the meta-atoms of the two lenses (Fig. 1d). If the phase change introduced by
the meta-atoms is local (i.e. the coupling between the meta-atoms is small), we can expect light scattered by each
group of meta-atoms (one group corresponds to one lens) to interfere constructively in their respective focal spot
at the design wavelength.
Metasurface structure.
We use a high contrast dielectric metasurface platform for implementation of
the devices. The platform consists of amorphous silicon (
α
-Si) nano-posts on a fused silica substrate (Fig. 2a)
that can form a hexagonal lattice (Fig. 2b). For proper choices of the nano-posts height and lattice constant,
full phase coverage can be achieved at a design wavelength by changing the diameters of the nano-posts
11
. The
nano-posts behave like multi-mode truncated waveguides with many resonant modes around the wavelength of
interest
18,
36
. Superposition of the scattered fields of these resonant modes can result in full 2
π
phase coverage,
T. J. Watson Laboratory of Applied Physics, California Institute of Technology, 1200 E. California Blvd., Pasadena, CA
91125, USA. Correspondence and requests for materials should be addressed to A.F. (email: faraon@caltech.edu)
r
eceived: 08 July 2016
a
ccepted: 15 August 2016
P
ublished: 06 September 2016
OPEN
www.nature.com/scientificreports/
2
SCIENTIFIC
RepoRts
| 6:32803 | DOI: 10.1038/srep32803
while maintaining a high transmission amplitude. Since the structure needs to be fabricated with a single step
electron beam lithography, the nano-post heights should be the same at both wavelengths (which we have chosen
to be 915 and 1550
nm due to availability of laser sources). In addition, since the two wavelengths are relatively
far apart, we choose the 1550
nm lattice constant to be twice that of the 915
nm (Fig. 3a,b). With this choice, the
two metasurfaces can be interleaved by simply replacing one out of four 915
nm meta-atoms by a 1550
nm one
(Fig. 3c).
Taking these considerations into account, we find that a post height of 718
nm, lattice constants of 360
nm at
λ
=
915 nm, and 720 nm at
λ
=
1550
nm, enable full phase coverage with high transmission at both wavelengths.
The simulated intensity transmission
t
()
2
and transmission phase
∠
t
()
for such uniform lattices are plotted in
Fig. 2d,e at 915 and 1550
nm, respectively. For simulations, a uniform array of meta-atoms with a given diameter
is illuminated with a plane wave at the wavelength of interest (Fig. 2c), and the transmission amplitude and phase
are calculated. We have used the rigorous coupled wave analysis
37
to perform the simulations.
Experimental Results
For experimental demonstration, we have designed and fabricated a multi-sector and an interleaved lens that
focus light from single mode fibers at 915 and 1550
nm to a focal point 400
μ
m away from the lens’ surface with-
out spherical aberrations. The lenses are 300
μ
m in diameter, and the single mode fiber is placed 600
μ
m away
from the backside of the ~500-
μ
m-thick substrate of the lens (resulting in a focal distance of 286
μ
m, and a
numerical aperture of 0.46). The multi-sector lens is formed by dividing two single wavelength lenses designed for
915 and 1550
nm to 8 radial sectors and combining them similar to Fig. 1c and the interleaved lens is formed from
combining the two single wavelength lenses in the manner shown in Fig. 3. The single wavelength lenses are
designed using the metasurface platforms described above. The smallest nano-post diameter is set to 72
nm in all
the designs. For the interleaved lens, a minimum gap of ~50
nm is set between the adjacent nano-posts to facilitate
their fabrication. This resulted in the maximum nano-post diameters of 200 and 420
nm for the 915 and 1550
nm
lenses, respectively, thus the highest achievable phase delay was ~1.6
π
at each wavelength. The less than 2
π
phase
coverage leads to small wavefront errors and lowers the focusing efficiencies of the lenses. In the design process,
the best nano-post for each lattice site was chosen by minimizing the complex transmission error defined as
a
b
c
d
Figure 1.
Spatial multiplexing scheme.
(
a
) Schematic of a metasurface lens designed to focus light with
wavelength
λ
1
and (
b
) wavelength
λ
2
to a distance
f
. (
c
) Double-wavelength metasurface lens formed by lens
aperture division, and (
d
) by interleaving meta-atoms.
www.nature.com/scientificreports/
3
SCIENTIFIC
RepoRts
| 6:32803 | DOI: 10.1038/srep32803
φ
−
expi
t
()
, where
φ
is the desired phase at the lattice site and
t
is the complex transmission coefficient of the
nano-posts. Using this design procedure, and assuming that the lenses require a uniform distribution of
nano-posts with various phases from 0 to 2
π
, we find that the incomplete phase coverage achieved here results in
a less than 3% reduction in the lens efficiency.
The devices were fabricated by depositing a 718-nm-thick layer of
α
-Si on a fused silica substrate using the
plasma enhanced chemical vapor deposition method. The device pattern was written on an electron beam resist
using e-beam lithography, and was transferred to an aluminum oxide layer using a lift-off process. The aluminum
oxide layer served as a hard mask for etching the
α
-Si layer in a dry etch process, and was removed in a solution
of hydrogen peroxide and ammonium hydroxide. Optical and scanning electron microscope images of both the
multi-sector and interleaved devices are shown in Fig. 4.
The lenses were characterized by measuring the intensity distributions in the focal plane, and in many planes
parallel to the focal plane using custom built microscopes with ~100
×
magnification. Schematics of measure
-
ment setups are shown in Fig. 5. Measured intensities in axial and focal planes at both wavelengths are plot-
ted in Fig. 6a–d for the multi-sector and in Fig. 6i–l for the interleaved lens. The polarization of incident light
100
140
180
220
0
1
]
m
n
[
r
e
t
e
m
a
i
D
]
m
n
[
r
e
t
e
m
a
i
D
Intensity transmission
Transmission phase/2
π
Intensity transmission
100
200
300
400
500
0
1
λ
=915 nm
λ
=1550 nm
Fused Silica
α
-Si
a
e
d
D
b
tE
α
-Si
Fused Silica
E
c
Transmission phase/2
π
a
Figure 2.
Metasurface structure with simulated amplitude and phase curves.
(
a
) Schematic of an amorphous
silicon (
α
-Si) cylindrical nano-post on a fused silica substrate. (
b
) Top view of the meta-atoms on a hexagonal
lattice showing geometrical parameters. (
c
) Schematic of the simulated structure. (
d
) Intensity transmission,
and phase of the transmission coefficient at
λ
=
915 nm and (
e
) at
λ
=
1550
nm. The lattice constant is 360
nm
for
λ
=
915
nm and 720
nm for the
λ
=
1550 nm structure. The
α
-Si layer is 718
nm thick in both cases.
a
a
b
2a
c
Figure 3.
Interleaved lattice schematics.
(
a
) Schematic of the short-wavelength metasurface lattice with a
lattice constant of
a
. (
b
) Schematic of the long-wavelength lattice with a lattice constant of 2
a
, overlaid on the
unit-cell boundaries of the short-wavelength meta-atoms. (
c
) Schematic of the interleaved lattice resulting from
replacing one out of four short-wavelength meta-atoms with a long-wavelength one.
www.nature.com/scientificreports/
4
SCIENTIFIC
RepoRts
| 6:32803 | DOI: 10.1038/srep32803
was changed using the polarization controllers shown in Fig. 5, and no polarization dependence was observed.
Measured axial plane intensities for the multi-sector lens are plotted in Fig. 6a,c, where a single strong focus is
observed at both wavelengths. The intensity distributions in the focal plane are plotted in Fig. 6b,d and show
features that are caused by the division of the lens aperture into multiple sectors. The high frequency fluctuations
observed in the 1550
nm focal plane measurements are caused by the highly non-uniform responsivity of the
phosphorous coated CCD used. To achieve smoother intensity distributions in the axial plane, these high fre-
quency fluctuations are filtered through removing all components with spatial frequencies higher than the free
space propagation constant. For comparison, a lens designed with the same method and with the same NA, but
with all dimensions and distances four times smaller than the fabricated device was simulated using the finite
difference time domain (FDTD) method in MEEP
38
. The smaller size of the simulated device was necessary to
make the simulations feasible with the available computational resources. Figure 6e–h show the simulated inten-
sities for this device at both wavelengths in the axial and focal planes. The illumination was linearly polarized in
simulations, and the symmetry of the structure ensures the same behavior for other incident polarizations. A
very good agreement is observed between simulated and measured focal depths and the focal plane intensities.
Focusing efficiency is defined as the ratio of the power focused by the device, to the output power of the fiber, and
is measured at 915 and 1550
nm with setups schematically shown in Fig. 5b,c, respectively. The pinhole used at
915
nm has a diameter of 20
μ
m, and the iris used for 1550
nm is 2
mm in diameter, which translates to a ~20
μ
m
diameter in the object plane. Simulated and measured efficiencies and full width at half maximums (FWHM),
in addition to the diffraction limited FWHMs are summarized in Tables 1 and 2, respectively. The diffraction
50
m
50
m
a
b
c
d
e
f
10
m
1
m
2
m
500 nm
Figure 4.
Fabricated device images.
(
a
) Optical and (
b
,
c
) scanning electron microscope images of the multi-
sector lens. (
d
) Optical and (
e
,
f
) scanning electron microscope images of the interleaved lens.
www.nature.com/scientificreports/
5
SCIENTIFIC
RepoRts
| 6:32803 | DOI: 10.1038/srep32803
limited FWHMs are found via simulation of a perfect phase mask at each wavelength, with the same optical fiber
illuminations that are used to design and measure the lenses.
Measurement results for the interleaved lens are plotted in Fig. 6i–l. Unlike 1550
nm, at 915
nm a second focus
is observed at
z
≈
220
μ
m, with the peak intensity approximately 1.8 times, and a power 1.2 times those of the
main focus. Similar to the multi-sector lens case, a four times smaller interleaved lens designed using the same
platform is simulated for comparison, and the simulated intensities are plotted in Fig. 6m–p. A weak second focus
is observed around
z
≈
50
μ
m in the simulations as well. It is worth noting that these devices are multi-order
(similar to multi-order gratings), and have multiple focal points (like a Fresnel zone plate lens). These higher
order focal points can be seen in all four axial plane measurements in Fig. 6a,c,i,k. The higher order focal points
have low intensities in all cases except for Fig. 6i. If the “
blazing
” of the lens is perfect (i.e. the phase profile is equal
to its ideal case), all of the power will be directed towards the designed focal distance. However, if some error is
introduced to the phase profile, a portion of the power will be directed towards higher order focal points. As this
error increases, the power in the main focus will decrease. The 1550
nm nano-posts are optically large and support
many resonant modes around 915
nm, resulting in some error in the phase of the total transmitted field at 915
nm.
Besides, as we will shortly discuss, the coupling between 1550 and 915
nm posts is not negligible at 915
nm. These
errors, result in a significant portion of the power going to the higher order focus at 915
nm for the interleaved
lens. Measured and simulated efficiencies and FWHMs are summarized in Tables 1 and 2, respectively.
To determine the origin of the focus observed at
z
≈
220
μ
m, we use paraxial imaging equations by consider
-
ing the fiber tip as an object and the focal plane intensity as its image. We find that the effective object distance
from the lens is 1003
μ
m using the
f
1
=
286
μ
m focal distance, and the 400
μ
m image distance. Therefore, the focal
distance corresponding to the focus observed at
z
≈
220
μ
m is
f
3
≈
180
μ
m. We represent this focal distance by
f
3
,
because there is also a secondary focal distance
f
2
=
485
μ
m arising from the 1550
nm nano-posts, as they also
form a lens at 915
nm. The focal distance of the lens formed from 1550
nm nano-posts is given by
μμ
=×
=
f
286
m
485
m
2
1550
915
at 915
nm
28
. We note that we have 1/
f
3
=
1/
f
1
+
1/
f
2
. This means that the equiv-
alent transmission mask of the lens contains a term proportional to
−+
+
π
λ
{}
()
ix
y
ex
p
ff
21
1
22
12
. This term
Objective lens
Tube lens
Camera
Device
a
Camera
Device
Pinhole
b
Optical power meter
Mirror
c
Optical power meter
Iris
Laser
Polarization
controller
Laser
Polarization
controller
Laser
Polarization
controller
Objective lens
Tube lens
Device
Figure 5.
Measurement setup schematics.
(
a
) Schematics of the measurement setup used for measuring
the optical intensity distribution patterns in different planes at 915
nm. (
b
) Schematics of the setup used for
measuring the focusing efficiency of the lens at 915
nm. (
c
) Schematics of the setup used for characterizing the
devices at 1550
nm. The flip mirror, iris, and optical power meter were used to measure the focusing efficiencies.
www.nature.com/scientificreports/
6
SCIENTIFIC
RepoRts
| 6:32803 | DOI: 10.1038/srep32803
is the result of addition of two phase profiles generated by the 915 and 1550
nm nano-posts, and indicates cou-
pling between these nano-posts (this term cannot exist if 915 and 1550
nm posts operate completely inde-
pendently). Therefore, at 915
nm the coupling between the interleaved nano-posts cannot be completely
neglected. For an optimal design, this coupling can be taken into account if unit cells formed from combining the
two nano-post groups are analyzed together
28
.
40
100
160
40
100
160
100
400
700
100
400
700
a
i
c
e
gh
f
b
mn
j
k
op
l
d
z[
μ
[
z
]
m
μ
m]
40 100 160
100
400
700
z[
μ
[
z
]
m
μ
m]
z[
μ
[
z
]
m
μ
m]
40 100 160
100
400
700
z[
μ
[
z
]
m
μ
m]
z
y
20
μ
m
z
y
20
μ
m
z
y
20
μ
m
z
y
20
μ
m
z
y
20
μ
m
z
y
20
μ
m
z
y
20
μ
m
λ
=915 nm
λ
=1550 nm
λ
=915 nm
z
y
20
μ
m
λ
=1550 nm
Measurement, Multi-sec
tor
Simulation, Multi-sec
tor
Measurement, Interlea
ve
dS
imulation, Interle
av
ed
1
0
Intensity [a.u.]
Figure 6.
Measured and simulated intensity profiles.
(
a
) Intensity measured in the axial plane and (
b
), in
the focal plane at 915
nm for the multi-sector lens. (
c
,
d
) Same measurements as (
a
,
b
) at 1550
nm. The high
frequency intensity fluctuations observed in the measured focal intensity at 1550
nm is caused by the high
non-uniformity in the responsivity of pixels in the phosphorous coated CCD used in 1550
nm measurements.
This nonphysical fluctuation is filtered for the axial plane plots to acquire a smoother distribution. (
e
–
h
) FDTD
simulated intensities for a similar multi-sector lens with a four times smaller size and focal distance. (
i
–
l
)
Similar results to (
a
–
d
), but for the interleaved lens. (
m
–
p
) FDTD simulated intensities for a four times smaller
interleaved lens. Scale bars: 4
μ
m.
Wavelength
915 nm
1550 nm
Measured
Simulated
Measured
Simulated
Multi-sector
37
±
1%
40.6%
30
±
1%
36.8%
Interleaved
10
±
0.5%
27%
58
±
1%
75.8%
Table 1.
Measured and simulated efficiencies for the multi-sector and interleaved lenses.
Wavelength
915 nm
1550 nm
Measured
Simulated
Limit
Measured
Simulated
Limit
Multi-sector
1.85
±
0.05
μ
m
1.7
μ
m
1.6
μ
m
3.3
±
0.2
μ
m
3
μ
m
2.75
μ
m
Interleaved
1.85
±
0.05
μ
m
1.7
μ
m
1.6
μ
m
3.3
±
0.2
μ
m
3
μ
m
2.75
μ
m
Table 2.
Measured, simulated, and diffraction limited FWHM focal spot sizes for the multi-sector and
interleaved lenses.
www.nature.com/scientificreports/
7
SCIENTIFIC
RepoRts
| 6:32803 | DOI: 10.1038/srep32803
Discussion
The efficiency of the interleaved lens at 915
nm is significantly lower than 1550
nm, both in measurements and
simulations. Two factors play important roles in this difference between the efficiencies at the two wavelengths.
First, the interleaved lens has an effective lattice constant of 720
nm (Fig. 3c), which is close to
λ
==
a
727
nm
n
ma
x
2
3
g
, the lattice constant above which higher diffraction orders will be propagating in the
glass substrate at 915
nm (here
n
g
=
1.452 is the refractive index of glass). Therefore, the non-periodicity of the
lens structure results in higher order diffractions propagating inside the substrate. In contrast, this lattice constant
is subwavelength enough at 1550
nm such that no higher diffraction orders are present. Second, 915
nm
nano-posts are optically small at 1550
nm, whereas 1550
nm nano-posts are optically large and support many
resonances around 915
nm. Therefore, while adding the 915
nm nano-posts to the 1550
nm lens results in a rela-
tively small phase error at 1550
nm, introducing the large 1550
nm nano-posts to the 915
nm lens changes the
phase profile significantly, and for some nano-post diameters, it also reduces the transmission amplitude. In
addition, the measured efficiency for the interleaved lens at 915
nm (10%) is lower than the simulated value of
27%. The transmission phase of the device at 915
nm is more sensitive to errors in the nano-post diameters
because of the larger aspect ratio of the nano-posts, and fabrication errors have degraded the phase profile of the
lens and its efficiency by directing a significant portion of power to the higher order focus around 220
μ
m.
Efficiencies of the multi-sector device at the two wavelengths are closer to each other than the interleaved
lens. The sum of simulated efficiencies at 915
nm and 1550
nm for these devices is always less than 100%. The
interleaved design, in contrary, can have a sum of efficiencies at 915 and 1550
nm higher than 100% as evidenced
by the simulation results. This is because the high index nano-posts can have an optical cross-section significantly
larger than the geometrical area of the metasurface pixel that they occupy. Besides, the efficiency of the inter
-
leaved lens can be increased, and its sensitivity to fabrication errors can be decreased using the meta-molecule
concept and a concurrent design of the nano-posts for the two wavelengths
28
. In addition, the division of the
aperture to multiple macroscopic sectors changes the shape of the input aperture of the lens and thus the shape of
its focal spot. The interleaved design on the other hand, does not cause this issue.
The demonstrated multiwavelength lenses can be used in applications where simultaneous operation at a few
discrete wavelengths is required, such as two photon fluorescence microscopy. While conventional refractive
achromatic lenses could have a better performance (specifically a higher efficiency) in such applications, they are
bulky, more expensive to fabricate, and harder to customize. Besides, multiple metasurfaces can be monolithically
integrated in order to correct various aberrations, add functionalities, or be directly integrated with electronics
to form compact electro-optical systems
39
. Nevertheless, multiwavelength lenses are demonstrated here only as
a proof of concept example for the general method of spatial multiplexing of metasurfaces for implementing
multiwavelength multifunctional optical devices. The introduced methods can be directly applied to designing
metasurfaces with different functionalities at different wavelengths. For instance, a metasurface can be designed
to operate as a lens at one wavelength, and as a grating at the other one. It can also be applied to making metasur
-
faces that perform multiple functions simultaneously at a single wavelength. It would be very challenging, if at all
possible, to fabricate such devices with the conventional refractive optics platform.
Conclusion
We have shown that by spatially multiplexing metasurface lenses that are designed for operation at two different
wavelengths, we can realize lenses that simultaneously operate at both wavelengths. We designed, fabricated, and
characterized double-wavelength lenses based on macroscopic aperture division (i.e. the multi-sector lens), and
meta-atom interleaving. Although here we used this concept to demonstrate double-wavelength lenses, the idea
can be readily generalized to devices with more operation wavelengths, or devices that perform different func-
tions at different wavelengths, or even at the same wavelength. Therefore, spatial multiplexing introduces a simple
route towards multiwavelength and multi-functional metasurfaces.
References
1.
Kildishev, A. V., Boltasseva, A. & Shalaev, V. M. Planar photonics with metasurfaces.
Science
339,
1232009 (2013).
2.
Yu, N. & Capasso, F. Flat optics with designer metasurfaces.
Nature Mater
.
13,
139–150 (2014).
3.
Jahani, S. & Jacob, Z. All-dielectric metamaterials.
Nature Nanotech
.
11,
23–36 (2016).
4.
Fattal, D., Li, J., Peng, Z., Fiorentino, M. & Beausoleil, R. G. Flat dielectric grating reflectors with focusing abilities.
Nature Photon
.
4,
466–470 (2010).
5.
Lu, F., Sedgwick, F. G., Karagodsky, V., Chase, C. & Chang-Hasnain, C. J. Planar high-numerical-aperture low-loss focusing
reflectors and lenses using subwavelength high contrast gratings.
Opt. Express
18,
12606–12614 (2010).
6.
Lalanne, P., Astilean, S., Chavel, P., Cambril, E. & Launois, H. Blazed binary subwavelength gratings with efficiencies larger than
those of conventional échelette gratings.
Opt. Lett.
23,
1081–1083 (1998).
7.
Astilean, S., Lalanne, P., Chavel, P., Cambril, E. & Launois, H. High-efficiency subwavelength diffractive element patterned in a high-
refractive-index material for 633
nm.
Opt. Lett.
23,
552–554 (1998).
8.
Lalanne, P., Astilean, S., Chavel, P., Cambril, E. & Launois, H. Design and fabrication of blazed binary diffractive elements with
sampling periods smaller than the structural cutoff.
J. Opt. Soc. Am. A
16,
1143–1156 (1999).
9.
Arbabi, A.
et al.
Controlling the phase front of optical fiber beams using high contrast metastructures. In
CLEO
:
2014
, OSA
Technical Digest (online), STu3M.4 (Optical Society of America, San Jose, California, 2014).
10.
Vo, S.
et al.
Sub-wavelength grating lenses with a twist.
IEEE Photonics Technol
.
Lett.
26,
1375–1378 (2014).
11.
Arbabi, A., Horie, Y., Ball, A. J., Bagheri, M. & Faraon, A. Subwavelength-thick lenses with high numerical apertures and large
efficiency based on high-contrast transmitarrays.
Nat. Commun.
6,
7069 (2015).
12.
Arbabi, A., Horie, Y., Bagheri, M. & Faraon, A. Dielectric metasurfaces for complete control of phase and polarization with
subwavelength spatial resolution and high transmission.
Nature Nanotech
.
10,
937–943 (2015).
13.
West, P. R.
et al.
All-dielectric subwavelength metasurface focusing lens.
Opt. Express
22,
26212–26221 (2014).
14.
Chong, K. E.
et al.
Polarization-independent silicon metadevices for efficient optical wavefront control.
Nano Lett.
15,
5369–5374
(2015).
www.nature.com/scientificreports/
8
SCIENTIFIC
RepoRts
| 6:32803 | DOI: 10.1038/srep32803
15.
Yu, Y. F.
et al.
High-transmission dielectric metasurface with 2
π
phase control at visible wavelengths.
Laser Photon. Rev.
9,
412–418
(2015).
16.
Arbabi, A., Briggs, R. M., Horie, Y., Bagheri, M. & Faraon, A. Efficient dielectric metasurface collimating lenses for mid-infrared
quantum cascade lasers.
Opt. Express
23,
33310–33317 (2015).
17.
Shalaev, M. I.
et al.
High-efficiency all-dielectric metasurfaces for ultracompact beam manipulation in transmission mode.
Nano
Lett.
15,
6261–6266 (2015).
18.
Kamali, S. M., Arbabi, A., Arbabi, E., Horie, Y. & Faraon, A. Decoupling optical function and geometrical form using conformal
flexible dielectric metasurfaces.
Nat. Commun.
7,
11618 (2016).
19.
Kamali, S. M., Arbabi, E., Arbabi, A., Horie, Y. & Faraon, A. Highly tunable elastic dielectric metasurface lenses.
arXiv
:
1604.03597
(2016).
20.
Faraon, A., Arbabi, A., Horie, Y., Arbabi, E. & Kamali, S. M. Flat free-space optical elements based on dielectric metasurfaces.
SPIE
Newsroom
6375 (2016).
21.
Zhan, A.
et al.
Low-contrast dielectric metasurface optics.
ACS Photonics
3,
209–214 (2016).
22.
Ren, Y.
et al.
Orbital angular momentum-based space division multiplexing for high-capacity underwater optical communications.
arXiv:1604.06865
(2016).
23.
Backlund, M. P.
et al.
Removing orientation-induced localization biases in single-molecule microscopy using a broadband
metasurface mask.
Nature Photon
.
10,
459–462 (2016).
24.
Gutruf, P.
et al.
Mechanically tunable dielectric resonator metasurfaces at visible frequencies.
ACS Nano
10,
133–141 (2016).
25.
Born, M. & Wolf, E.
Principles of Optics: Electromagnetic Theory of Propagation, Interference and Diffraction of Light
(Cambridge
University Press, 1999).
26.
O’Shea, D. C., Suleski, T. J., Kathman, A. D. & Prather, D. W.
Diffractive Optics: Design, Fabrication and Test
(SPIE Press, 2004).
27.
Aieta, F., Kats, M. A., Genevet, P. & Capasso, F. Multiwavelength achromatic metasurfaces by dispersive phase compensation.
Science
347,
1342–1345 (2015).
28.
Arbabi, E., Arbabi, A., Kamali, S. M., Horie, Y. & Faraon, A. Multiwavelength polarization-insensitive lenses based on dielectric
metasurfaces with meta-molecules.
Optica
3,
628–633 (2016).
29.
Khorasaninejad, M.
et al.
Achromatic metasurface lens at telecommunication wavelengths.
Nano Lett.
15,
5358
‒
5362 (2015).
30.
Eisenbach, O., Avayu, O., Ditcovski, R. & Ellenbogen, T. Metasurfaces based dual wavelength diffractive lenses.
Opt. Express
23,
3928–3936 (2015).
31.
Zhao, Z.
et al.
Multispectral optical metasurfaces enabled by achromatic phase transition.
Sci. Rep.
5,
15781 (2015).
32.
Arbabi, E., Arbabi, A., Kamali, S. M., Horie, Y. & Faraon, A. High efficiency double-wavelength dielectric metasurface lenses with
dichroic birefringent meta-atoms.
Opt. Express
24,
18468–18477 (2016).
33.
Bayer, B. Color imaging array.
US Patent
3,971,065 (1976).
34.
Zhao, W.
et al.
Full-color hologram using spatial multiplexing of dielectric metasurface.
Opt. Lett.
41,
147–150 (2016).
35.
Maguid, E.
et al.
Photonic spin-controlled multifunctional shared-aperture antenna array.
Science
352
, 1202–1206 (2016 ).
36.
Lalanne, P. Waveguiding in blazed-binary diffractive elements.
J. Opt. Soc. Am. A
16,
2517–2520 (1999).
37.
Liu, V. & Fan, S. S4: A free electromagnetic solver for layered periodic structures.
Comput. Phys. Commun.
183,
2233–2244 (2012).
38.
Oskooi, A. F.
et al.
Meep: A flexible free-software package for electromagnetic simulations by the fdtd method.
Comput. Phys.
Commun.
181,
687–702 (2010).
39.
Arbabi, A.
et al.
An optical metasurface planar camera.
arXiv:1604.06160
(2016).
Acknowledgements
This work was supported by Samsung Electronics. A.A. was also supported by National Science Foundation
award 1512266. S.M.K. was supported as part of the Department of Energy (DOE) “Light-Material Interactions
in Energy Conversion Energy Frontier Research Center” under grant no. DE-SC0001293. The devices were
fabricated at the Kavli Nanoscience Institute at Caltech.
Author Contributions
E.A., A.A. and A.F. conceived the experiments. E.A., A.A., S.M.K. and Y.H. performed the simulations and
fabricated the devices. E.A. performed the measurements, and analyzed the data. E.A., A.A. and A.F. co-wrote the
manuscript. All authors discussed the results and commented on the manuscript.
Additional Information
Competing financial interests:
The authors declare no competing financial interests.
How to cite this article
: Arbabi, E.
et al.
Multiwavelength metasurfaces through spatial multiplexing.
Sci. Rep.
6
,
32803; doi: 10.1038/srep32803 (2016).
This work is licensed under a Creative Commons Attribution 4.0 International License. The images
or other third party material in this article are included in the article’s Creative Commons license,
unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license,
users will need to obtain permission from the license holder to reproduce the material. To view a copy of this
license, visit http://creativecommons.org/licenses/by/4.0/
© The Author(s) 2016