Controllable optical activity with non-chiral plasmonic metasurfaces

Abstract

Optical activity is the rotation of the plane of linearly polarized light along the propagation direction as the light travels through optically active materials. In existing methods, the strength of the optical activity is determined by the chirality of the materials, which is difcult to control quantitatively. Here we numerically and experimentally investigated an alternative approach to realize and control the optical activity with non-chiral plasmonic metasurfaces. Through judicious design of the structural units of the metasurfaces, the right and left circular polarization components of the linearly polarized light have different phase retardations after transmitting through the metasurfaces, leading to large optical activity. Moreover, the strength of the optical activity can be easily and accurately tuned by directly adjusting the phase difference. The proposed approach based on non-chiral plasmonic metasurfaces exhibits large optical activity with a high controllable degree of freedom, which may provide more possibilities for applications in photonics.

OPEN

Light: Science & Applications (2016) 5, e16096; doi:http://dx.doi.org/10.1038/lsa.2016.96

Web End =10.1038/lsa.2016.96

& 2016 CIOMP. All rights reserved 2047-7538/16 http://www.nature.com/lsa

Web End =www.nature.com/lsa

Ping Yu1, Jianxiong Li1, Chengchun Tang2, Hua Cheng1, Zhaocheng Liu1, Zhancheng Li1, Zhe Liu2, Changzhi Gu2, Junjie Li2, Shuqi Chen1 and Jianguo Tian1

Optical activity is the rotation of the plane of linearly polarized light along the propagation direction as the light travels through optically active materials. In existing methods, the strength of the optical activity is determined by the chirality of the materials, which is difcult to control quantitatively. Here we numerically and experimentally investigated an alternative approach to realize and control the optical activity with non-chiral plasmonic metasurfaces. Through judicious design of the structural units of the metasurfaces, the right and left circular polarization components of the linearly polarized light have different phase retardations after transmitting through the metasurfaces, leading to large optical activity. Moreover, the strength of the optical activity can be easily and accurately tuned by directly adjusting the phase difference. The proposed approach based on non-chiral plasmonic metasurfaces exhibits large optical activity with a high controllable degree of freedom, which may provide more possibilities for applications in photonics.

Light: Science & Applications (2016) 5, e16096; doi:http://dx.doi.org/10.1038/lsa.2016.96

Web End =10.1038/lsa.2016.96; published online 1 July 2016

Keywords: metasurfaces; optical activity; surface plasmon polaritons

INTRODUCTION

Optical activity is the rotation of the plane of linearly polarized (LP) light along the propagation direction as the light travels through optically active materials, and it has acquired considerable importance in analytical chemistry, spectroscopy, crystallography and molecular biology1,2. Optical activity is a type of birefringence, and it occurs in solutions of chiral molecules such as sucrose, solids with rotated crystal planes such as quartz, and spin-polarized gases of atoms or molecules. The molecular structures of these materials exhibit a geometry where the original structure cannot be made congruent with its mirror image3. When LP light is transmitted through optically active materials, the right and left circular polarization components of the LP light experience different refractive indexes, leading to a change in the relative phase between the two circular polarizations. The polarization of the transmitted light is thus rotated at an angle4,5. The

degree of rotation depends on the propagation length in the material and the specic rotation, which is proportional to the refractive index difference (n). For conventional optically active materials, n is related to the specic crystal system of the birefringent crystal.

Therefore, the strength of optical activity can be controlled by tuning the molecule arrangements of special optically active materials, such as liquid crystals and magneto-optical crystals.

Metamaterials are articial media on a size scale smaller than the wavelength of incident light, and the articial structural units function as atoms and molecules in conventional materials68. Through

ORIGINAL ARTICLE

Controllable optical activity with non-chiral plasmonic metasurfaces

regulated interactions with electromagnetic waves, metamaterials can produce remarkable physical properties unavailable in naturally occurring or chemically synthesized materials913. By designing structural units with chirality, metamaterials can also generate large optical activity with a mechanism similar to that in traditional optically active materials1418. Recently, plasmonic metasurfaces have been introduced, which are metamaterials with reduced dimensionality, typically consisting of a two-dimensional arrays of nanostructures1924.

Plasmonic metasurfaces are capable of controlling the optical phase of transmitted or reected light at the nanoscale and generating a wide range of position-dependent discontinuous interfacial phase proles.Research interests have moved toward fascinating phenomena and applications, such as anomalous refraction or reection19,2528, vortex

or vector beam generation2931 and wave plates3238. Because plasmonic metasurfaces enable unprecedented degrees of freedom in light phase manipulation, they can be used to directly design the transmitted phase of right circularly polarized (RCP) and left circularly polarized (LCP) light. Optical activity can thus also be realized under normal incidence through another efcient method in which the chirality of the nanostructures is not a critical factor. Therefore, metasurfaces provide a feasible method for realizing and controlling optical activity via directly designing atoms or molecules in nonchiral plasmonic metasurfaces.

Here we propose an alternative approach to realize controllable optical activity using single-layer, non-chiral, plasmonic metasurfaces

1The MOE Key Laboratory of Weak Light Nonlinear Photonics, School of Physics, Teda Applied Physics Institute, and the 2011 Project Collaborative Innovation Center for

Biological Therapy, Nankai University, Tianjin 300071, China and 2Beijing National Laboratory for Condensed Matter Physics, Institute of Physics, Chinese Academy of Sciences,

Beijing 100190, China

Correspondence: SQ Chen, Email: mailto:[email protected]

Web End [email protected] ; JG Tian, Email: mailto:[email protected]

Web End [email protected] ; JJ Li, Email: mailto:[email protected]

Web End [email protected]

Received 30 November 2015; revised 21 February 2016; accepted 21 February 2016; accepted article preview online 22 February 2016

Controllable optical activity with metasurfaces

P Yu et al

2

to analyze the polarization status of the transmitted light. Optical transmission was collected by an optical spectrum analyzer via a ber coupler, which enabled verication of the polarization rotation phenomenon generated by the sample. By rotating the angle of the analyzer within a range of 180, we could obtain the polarization state of the transmitted light over a broad wavelength range. All optical elements, including the microscope objective, lens, polarizers and detector, were operated in the broadband range.

Sample fabricationSputtering deposition, electron-beam lithography and reactive-ion etching were used to fabricate the metallic structures. The samples for generating different optical rotation were fabricated as follows. Silver was deposited onto the quartz plate to a thickness of 35 nm using a radio-frequency magnetic sputtering system, and a 100-nm-thick PMMA resist was subsequently spin-coated onto the sample. The sample was then subjected to bakeout at 120 C on a hotplate for 1 min. The cross-shaped pattern was exposed by the direct line writing method with an electron-beam lithography system (JEOL, JBX6300FS) at 100 keV and 3000 C cm1. After exposure, the sample was developed in MIBK:IPA (1:3) for 40 s and IPA for 30 s and was then blown dry using pure nitrogen. Finally, the pattern was transferred onto the silver layer by Ar plasma reactive-ion etching at a power of 100 W, a ow rate of 40 sccm and a pressure of 30 mTorr. The remaining PMMA resist was removed using hot acetone at 60 C for 30 min.

RESULTS AND DISCUSSIONTheoretical analysisAny LP light can be written as an equal combination of RCP and LCP light40:

ELP ERCP eifELCP 1 where E is the electric eld of the light. The phase difference

between the two circular polarizations governs the direction of the linear polarization. The optical activity is the rotation of LP light by adjusting . In contrast to existing optically active materials with chirality, the non-chiral plasmonic metasurface employed in our novel method enables direct modulation of the phase difference and thus enables controllable optical activity. The process of optical activity generation involves two steps: (i) conversion of the LP incident light into RCP and LCP beams and (ii) modulation of the phase difference of the LCP and RCP beams, and generation of new LP light with rotated plane due to local coupling.

We rst focus on the generation of the RCP and LCP beams. The Dirac bracket notation is adopted to dene the light polarization state. When LP light J

j i 1 1

Tis normally incident on the non-chiral

metasurface, the interaction between the light and the metasurface can be expressed as41,42

P J

^ j i

^

Px Jxj i

composed of cross-shaped nanoaperture arrays. Through accurate design of the structural units of the non-chiral metasurfaces, the incident LP light can be converted into LCP and RCP beams on a nanoscale. The phase retardation difference between the LCP and RCP beams can be easily modulated by varying the geometrical parameters of the nanoapertures, leading to continuously controllable optical activity. We theoretically analyzed the physical mechanisms and numerically and experimentally proved the theory using the designed non-chiral metasurfaces. Compared with metamaterials with chirality, the proposed non-chiral metasurfaces could more directly manipulate the phase retardation, which provides another efcient method for realizing optical activity.

MATERIALS AND METHODSNumerical simulationsAll numerical simulations were performed using the commercial software COMSOL Multiphysics based on the nite element method. The

permittivity of silver is dened by the Drude model Ag

0 N

f 2p

f f ig

with N 5, fp = 2.175 PHz and = 4.35 THz39.

We chose silver because it performs well in manipulating the light phase compared with other metals, such as gold. The refractive index of the glass substrate is taken as 1.47. For the theoretical simulations, the unit cell has periodic boundary conditions in the x and y planes, and the waveguide ports boundary conditions on other boundaries.

Measurement setup for optical activityThe experiments were performed using a custom-built optical setting, as shown in Figure 1. White light from a bromine tungsten lamp in the normal direction was used as the light source. The spot size of the incident light needs to be adjusted according to the size of the samples in the experimental process. Two lenses form a 4f-system. The aperture is set near the focus of the 4f-system. The sample can be sufciently illuminated by the incident light by adjusting the size of the aperture. A polarizer (P1) was used to control the normal incident LP light polarized along = 45, which was focused on the sample with a reective objective (RO1). The transmission through the sample was collected using another similar reective objective (RO2). Another polarizer (P2) was inserted into the optical path

OSA

P2

RO2

Sample

RO1

P1

Lens2

Aperture

Lens1

^

Py Jy

BTL

2

where Jx

j i 1 0

T and Jy

0 1

T are two orthogonal

components in the x and y directions of the incident LP light, respectively. The operator ^

P represents the inuence of the metasur-face on the amplitude transmission and phase change of the incident light. The two orthogonal components of ^

P can be expressed as

^

Px Jxj i Tx Jx

j ieij

x

Figure 1 White light is generated by a BTL. The polarization direction of incident light is controlled by polarizer P1. As the angle of another polarizer P2 is rotated, the transmission through the sample is collected by the OSA. Two reective objectives (RO1 and RO2) are used to focus and collect the light. BTL, bromine tungsten lamp; OSA, optical spectrum analyzer.

3

^

Py Jy

Ty Jy

eij

y

4

Light: Science & Applications doi:http://dx.doi.org/10.1038/lsa.2016.96

Web End =10.1038/lsa.2016.96

Controllable optical activity with metasurfaces

P Yu et al

3

where Tm and eim (m = x, y) denote the amplitude transmittance and phase factor of the transmitted beams in the m direction. We replace the LP light with two opposite-helicity circularly polarized (CP) beams to simplify the mechanism analysis. By substituting Equations (3) and(4) into Equation (2) and applying the replacement Jx

j i

1

2 R

j i L

j i

and Jy

iF Lj i 6

Assuming the transmission amplitudes of RCP and LCP waves have the same value TxR = TxL, we can obtain the transmitted light with a simple form, as follows:

TxReijxR Rj i TxLeij

xL j i TxReij

xR

L

j i TxLeij

xR

R

12i R

j i L

j i

, Equation (2) can be rewritten as

^ j i

1

2 Txeij

R

j i

1 iTyeij

1

2 Txeij

x

1 iTyeij

L

j i

xL exp

i2 jxR jxL

7

Equation (7) shows that the transmitted light exhibits linear polarization, and the polarization direction is governed by /2.

Sub-units of non-chiral metasurfacesBased on the characteristics of exible phase modulation, the process described above for optical activity generation can be achieved using non-chiral plasmonic metasurfaces. As shown in Figure 2a, the controllable optical activity arises from the phase control of the right and left circular polarization components decomposed from the LP light. Compared with the conventional method of achieving optical

sin F

2

x

y

L

j i 5

y

where R

j i 1 i

T and L

j i 1 i

2TxR

cos F

2

T correspond to the RCP and

LCP beams, respectively. According to Equation (5), if Tx = Ty, RCP or LCP light can be obtained when the phase difference satises jx

jy p=2 or /2. Thus, we can convert the LP incident light into

RCP or LCP light by designing different nanostructures. The expressions of two generated CP beams can be simplied as TxReijxR Rj i and

TxLeijxL Lj i, where Txn and xn (n = R, L) represent the transmission

amplitude and phase of the RCP and LCP beams along the x-axis.

The LCP and RCP beams need to be modulated to generate a phase difference F jxL jxR before they are superposed into new LP

light. The combination process of the RCP and LCP beams can be

written as:

TxReijxR Rj i TxLeij

a b

350

300

250

200

150

150 200 250 300

ly (nm)

ly

1.0

Non-chiral

metasurface

[afii9835]+x

Einc

0.5

Cross-shaped apertures

Birefringent materials

lx

R

R L

Eout

Eout

Eout

L

l x(nm)l x(nm)

0.0

0.5

Chiral structures

S3

350

1.0

c d

350

0.8

/4

0

0.6

300

250

3/4

l x(nm)

0.4

0.2

350

300

250

200

150

/2

200

T [afii9850]x

150

0.0

150

200 200

250 250

300 300

350 350

150

ly (nm)

ly (nm)

Figure 2 (a) Theory of controllable optical activity (middle region). Top and bottom regions indicate the optical rotations with non-chiral plasmonic metasurfaces and conventional methods, respectively. is the angle between the incident light and the x-axis. (b) Normalized Stokes parameter S3, (c) transmission T and (d) phase x in the x-direction as functions of lx and ly, which correspond to the lengths of the cross-shaped nanoaperture, as shown in the inset of b. The periods in the x and y directions are lx+100 nm and ly+100 nm, respectively.

doi:http://dx.doi.org/10.1038/lsa.2016.96

Web End =10.1038/lsa.2016.96

Light: Science & Applications

Controllable optical activity with metasurfaces

P Yu et al

4

activity using birefringent materials, we utilized a cross-shaped nanoaperture as the sub-unit because it can be easily designed for the conversion of LP light into CP light43. Two different types of cross-shaped nanoapertures that could generate RCP and LCP light were combined, thus forming a unit cell of the proposed non-chiral plasmonic metasurface. The relative phase retardation between two opposite-helicity CP beams can be tuned by directly changing the geometrical parameters of the nanoapertures while removing the additional external conditions, such as the external electric or magnetic elds. It is therefore possible to simultaneously realize and control the optical activity by designing the geometrical parameters of the two sub-units of the cross-shaped nanoapertures.

We now aim at tuning the lengths of the cross-shaped nanoaperture to nd the specic sub-units that can satisfy the requirements of the theory. A 35-nm-thick silver lm with cross-shaped nanoapertures is illuminated by the LP incident plane wave at a wavelength of 990 nm and exhibits polarization along the direction = 45. The corresponding sub-unit is shown in the inset of Figure 2b. The theoretical simulations are studied by the nite element method using COMSOL Multiphysics (see the Materials and Methods section)44. As mentioned above, the rst step of realizing optical activity is designing the nanostructures for the conversion of LP light into RCP and LCP light. Figure 2b illustrates the normalized Stokes parameter S3 as a function of the two lengths lx and ly of the cross-shaped nanoaperture. A value of S3 close to 1 or 1 represents nearly perfect RCP or LCP light, respectively, as shown by the red and purple regions in Figure 2b. Two types of sub-units of the non-chiral metasurface should therefore be selected from each region. When placing two sub-units close together, the output LP light is achieved at the far eld as a result of the local coupling between the generated RCP and LCP light. The next key issue is the generation of LP light with desired optical activity. Two modulated CP lights with opposite-handedness must have similar amplitude with a relative phase delay . Figure 2c and 2d shows the transmission T and phase x in the x-direction of cross-shaped nanoapertures. As illustrated in Figure 2c, the nanoapertures in the region of gray lines have similar T values of ~ 45%. The phase retardation between the LCP and RCP light can be modulated because the transmitted phase strongly depends on the geometrical parameters of the nanoaperture, as shown in Figure 2d. Considering the above results, the sub-units that satisfy the requirements of the theory are indicated by two dashed lines in Figure 2d. Rj and Lj are used to represent the sub-units that generate the RCP and LCP light, where j is the index of the sub-unit. To simplify the process, we x the geometrical parameters of sub-unit R1, indicated by the black square, and vary the other sub-unit Lj along the white curve in Figure 2d to vary the phase retardation from /2 to . Thus, new LP light with a different degree of optical activity can be realized.

Process and characteristics of the controllable optical activity Figure 3 shows the process of realizing the optical activity of LP incident light. A non-chiral plasmonic metasurface is proposed in which each unit cell consists of two sub-units of cross-shaped nanoapertures. These two sub-units split the incident LP beam into two CP beams with opposite-helicity in the near eld. These two rays effectively generate new LP light whose polarization has a different direction to that of the incident light. The unit cell of the presented non-chiral metasurface and its parameters are shown in the inset of Figure 3. Due to the coupling between the two sub-units, the periodic lengths of the unit cell have been slightly modied compared with those shown in Figure 2. As mentioned above, we x the geometrical parameters of the upper cross-shaped nanoaperture R1 while varying

the lx2 and ly2 of the lower cross-shaped nanoaperture Lj along the white curve shown in Figure 2d.

The calculated optical rotations and the corresponding transmissions at a wavelength of 990 nm are illustrated in Figure 4 when combing R1 and different Lj. The degree of the optical activity effect changes nonlinearly with variation in the geometrical parameters of Lj.

The variation in the geometrical parameters generates a relative phase difference from ~ /2 to . According to Equation (7), optical rotation with arbitrary values from 3 to 42 is obtained. We thus achieved controllable optical activity with relatively high-amplitude transmission beyond 40%. In addition, the transmitted light through each unit cell has high-quality linear polarization. To illustrate this point, we chose four specic structures to calculate the polarization patterns of the transmitted beams, as shown in the insets of Figure 4.

ly1

l

ly2 l

y x

z

Figure 3 Sketch of the optical activity process. The dashed line in the inset shows the unit cell of the presented non-chiral plasmonic metasurfaces designed with lx1 = 250 nm and ly1 = 200 nm; the periods in the x and y directions are 295 and 680 nm, and all nanoapertures have the same width of 43 nm.

36

DoLP=0.972

Rotations ()

48 1.0

0.8

0.6

0.4

0.2

0.0

Transmission

24

12

0

DoLP=0.969 DoLP=0.931

DoLP=0.999

250 260 270 280 290 300 310 320 330

ly2 (nm)

Figure 4 Calculated optical rotations (red line) and amplitude transmissions (blue line) at a wavelength of 990 nm. The parameter ly2 represents the subunit Lj along the white curve shown in Figure 2d. The insets show the polarization states and the corresponding DoLP for four metasurfaces, which correspond to the combinations of R1 and L1 to L4.

Light: Science & Applications doi:http://dx.doi.org/10.1038/lsa.2016.96

Web End =10.1038/lsa.2016.96

Controllable optical activity with metasurfaces

P Yu et al

5

The four selected non-chiral metasurfaces correspond to the combinations of sub-unit R1 and sub-units L1 to L4, which are also marked by colorized patterns in the white curve shown in Figure 2d. All of the resultant polarizations are found to be almost linear. The degree of

linear polarization (DoLP), dened as

S21S22

p

S0 , which is represented using Stokes parameters, is also given to effectively show the linear polarization of the transmitted wave. Therefore, the degree of optical activity can be exibly controlled by appropriately designing the nonchiral plasmonic metasurface without changing the incident wavelength.

To further analyze the properties of the proposed non-chiral metasurface, we calculated the DoLP and the rotations in the broadband range of the non-chiral metasurface with a combination of R1 and L4, as shown in Figure 5a and 5b. A DoLP value above 0.9 indicated by the shadowed region can be regarded as perfect LP light.

The polarization direction of the LP light has an optical rotation of 42 at the far eld within the shadowed region. The corresponding normalized Stokes parameter S3 distribution at 5 nm from the exit surface is shown in Figure 5c. The RCP light is obtained from the upper sub-unit, whereas the LCP light is obtained from the lower sub-unit at the near eld. As expected, the results verify the proposed theory that the RCP and LCP light are individually generated at the near eld and that new LP light is effectively generated at the far eld.

Measurement of controllable optical activityExperiments were performed to demonstrate the polarization responses of the designed non-chiral plasmonic metasurfaces. In the experimental process, a polarizer was used to control the polarization direction of the normal incident LP light along = 45. After passing through the non-chiral metasurfaces, the original polarization direction of incident light is rotated based on different phase delays between the LCP and RCP light. Another polarizer is used to analyze the polarization state of the transmitted light (see the Materials and Methods section). Figure 6a-6d shows the scanning electron microscopy images of the four non-chiral metasurface samples corresponding to combinations of R1

and L1 to L4. These congurations are experimentally achieved by sputtering deposition of silver lm and silicon dioxide and then applying electron-beam lithography followed by reactive-ion etching (see the Materials and Methods section). Figure 6e shows the obtained amplitude transmissions as a function of the analyzer angle while maintaining = 45. The observed optical rotations , which are characterized by the angle between and , are 3, 9, 23 and 30 for the four samples. All the angles are dened relative to the x-axis. We experimentally realized the controllable optical activity at the resonant wavelength of 970 nm by designing different non-chiral metasurfaces. The experimental results are in good agreement with the corresponding numerical results shown in Figure 6f. The slight differences between the experimental and simulated results, such as the degree of rotation and the decrease in the amplitude of transmission, may be caused by the discrepancy of the aperture lengths or the rounding of aperture corners in the fabrication process. In contrast to previous works using intrinsically non-chiral nanostructures (superimposed with their mirror images) under oblique incidence4548,

our proposed method realizes the optical activity with the non-chiral metasurface illuminated by normal incidence.

The experimentally measured transmitted intensity at the broad wavelength range of sample 4 is also shown in Figure 7a. The proposed optical activity with 30 rotation of LP light can be observed at the wavelength range from ~ 925 to 1025 nm. We also numerically calculated the corresponding transmitted intensity as a function of the analyzer angle , as shown in Figure 7b. The bandwidth of the intensity is almost the same as the corresponding experimental results. The difference is that the experimental resonate wavelength range exhibits a blueshift compared with the simulated results, which may be caused by the thickness difference of the glass substrate between the actual fabrication and the simulations. The proposed controllable optical activity effect can thus also be observed in a broadband wavelength range both numerically and experimentally, which could provide some opportunities in the design of various broadband optical systems and applications.

General applicability of the controllable optical activityThe proposed approach is not limited only by a certain wavelength range or nanostructure, which has a more general applicability. It is possible to realize a similar optical activity effect at other spectral regimes by changing the geometric parameters or designing other nanostructures. The key factor is that the nanostructures could be easily designed for converting LP light into LCP and RCP light. Signicant optical activity can also be achieved by using other non-chiral plasmonic metasurfaces based on the same method. For instance, similar controllable optical activity can be achieved by designing a non-chiral metasurface composed of cross-shaped nanoantennas (see Supplementary Fig. S1 and Supplementary Note 1); the controllable range of optical activity can also be further extended by using a dual-layer, non-chiral, plasmonic metasurface (see Supplementary Figs. S2 and S3 and Supplementary Note 2).

CONCLUSIONSIn conclusion, we present a novel method for realizing optical activity using an ultrathin, non-chiral, plasmonic metasurface consisting of cross-shaped nanoapertures. The physical mechanisms have been analyzed in detail and are different from those of traditional methods that realize optical activity using structures with chirality. We generate the phase delay between LCP and RCP beams by directly designing two different structural units. The optical activity can be exibly controlled by appropriately varying the non-chiral, optically active, articial nanostructures. In addition, the controllable property of the

a

c

1.0

DoLP Rotations ()

b

1.00.80.60.40.20.0

40 30 20 10

0 950 1000 1050 1100

1.0

0.0

Wavelength (nm)

Figure 5 Calculated far-eld (a) DoLP and (b) rotations for the non-chiral metasurface by the combination of R1 and L4 900 nm from the exit surface.

(c) Corresponding near-eld normalized Stokes parameters S3 5 nm from the exit surface at a wavelength of 990 nm.

doi:http://dx.doi.org/10.1038/lsa.2016.96

Web End =10.1038/lsa.2016.96

Light: Science & Applications

Controllable optical activity with metasurfaces

P Yu et al

6

a

e

90

0.4

120

60

Experiment

Simulation

150

30

Transmission Transmission

0.80.60.40.20.00.20.40.60.8

b

0.30.20.10.00.10.20.30.4

500 nm

180

0

330

300 Sample 1

Sample 2

Sample 3

Sample 4

210

270

c

240

120

f

90

60

150

30

d

180

0

210

330

240

300

270

Figure 6 (ad) SEM images of the four fabricated non-chiral plasmonic metasurfaces from sample 1 to sample 4, which correspond to the nanostructures by combination of R1 and L1 to L4. (e) Experimental and (f) numerically simulated transmissions as a function of analyzer angle for the four samples.

a b

180

Analyzer angle ()

135

90

0.21 0.45

0.34

0.23

0.11

0.00

180

135

90

45

0

Experiment

Simulation

0.16

0.11

Analyzer angle ()

(75)

45

0.06

0.00

0 850 900 950 1000 1050 950 1000 1100

1050

Wavelength (nm) Wavelength (nm)

Figure 7 (a) Experimentally and (b) numerically simulated transmissions as a function of analyzer angle at a broad wavelength range for sample 4.

optical activity is numerically and experimentally proved, which could provide some potential applications in micro-optical devices and sensors.

ACKNOWLEDGEMENTSThis work was supported by the National Basic Research Program (973 Pro

gram) of China (2012CB921900), the Chinese National Key Basic Research

Special Fund (2011CB922003), the Natural Science Foundation of China

(11574163, 61378006, 11304163 and 91323304), the Program for New Century

Excellent Talents in University (NCET-13-0294), the 111 project (B07013), and

the National Science Fund for Talent Training in Basic Sciences (J1103208).

1 Kuzmenko AM, Shuvaev A, Dziom V, Pimenov A, Schiebl M et al. Giant gigahertz optical activity in multiferroic ferroborate. Phys Rev B 2014; 89: 174407.

2 Park HS, Kim TT, Kim HD, Kim K, Min B. Nondispersive optical activity of meshed helical metamaterials. Nat Commun 2014; 5: 5435.

Light: Science & Applications doi:http://dx.doi.org/10.1038/lsa.2016.96

Web End =10.1038/lsa.2016.96

Controllable optical activity with metasurfaces

P Yu et al

7

3 Barron LD. Molecular Light Scattering and Optical Activity. Cambridge: Cambridge University Press; 2004.

4 Ogawa N, Seki S, Tokura Y. Ultrafast optical excitation of magnetic skyrmions. Sci Rep 2015; 5: 9552.

5 Okamura Y, Kagawa F, Seki S, Kubota M, Kawasaki M et al. Microwave magnetochiral dichroism in the chiral-lattice magnet Cu2OSeO3. Phys Rev Lett 2015; 114: 197202.

6 Zheludev NI. The road ahead for metamaterials. Science 2010; 328: 582583.7 Liu YM, Zhang X. Metamaterials: a new frontier of science and technology. Chem Soc Rev 2011; 40: 24942507.8 Zheludev NI. Obtaining optical properties on demand. Science 2015; 348: 973974.9 Shelby RA, Smith DR, Schultz S. Experimental verication of a negative index of refraction. Science 2001; 292: 7779.

10 Valentine J, Zhang S, Zentgraf T, Ulin-Avila E, Genov DA et al. Three-dimensional optical metamaterial with a negative refractive index. Nature 2008; 455: 376379.

11 Nguyen VC, Chen L, Halterman K. Total transmission and total reection by zero index metamaterials with defects. Phys Rev Lett 2010; 105: 233908.

12 Liu N, Wen FF, Zhao Y, Wang YM, Nordlander P et al. Individual nanoantennas loaded with three-dimensional optical nanocircuits. Nano Lett 2013; 13: 142147.

13 Cui YH, Kang L, Lan SF, Rodrigues S, Cai WS. Giant chiral optical response from a twisted-arc metamaterial. Nano Lett 2014; 14: 10211025.

14 Kuwata-Gonokami M, Saito N, Ino Y, Kauranen M, Jemovs K et al. Giant optical activity in quasi-two-dimensional planar nanostructures. Phys Rev Lett 2005; 95: 227401.

15 Bai B, Svirko Y, Turunen J, Vallius T. Optical activity in planar chiral metamaterials: theoretical study. Phys Rev A 2007; 76: 023811.

16 Yin XH, Schferling M, Metzger B, Giessen H. Interpreting chiral nanophotonic spectra: the plasmonic Born-Kuhn model. Nano Lett 2013; 13: 62386243.

17 Li ZF, Mutlu M, Ozbay E. Chiral metamaterials: from optical activity and negative refractive index to asymmetric transmission. J Opt 2013; 15: 023001.

18 Kang L, Lan SF, Cui YH, Rodrigues SP, Liu YM et al. An active metamaterial platform for chiral responsive optoelectronics. Adv Mater 2015; 27: 43774383.

19 Yu NF, Genevet P, Kats MA, Aieta F, Tetienne JP et al. Light propagation with phase discontinuities: generalized laws of reection and refraction. Science 2011; 334: 333337.

20 Yu NF, Capasso F. Flat optics with designer metasurfaces. Nat Mater 2014; 13: 139150.21 Cheng H, Liu ZC, Chen SQ, Tian JG. Emergent functionality and controllability in few-layer metasurfaces. Adv Mater 2015; 27: 54105421.

22 Yin XB, Ye ZL, Rho J, Wang Y, Zhang X. Photonic spin Hall effect at metasurfaces. Science 2013; 339: 14051407.

23 Li T, Liu H, Wang SM, Yin XG, Wang FM et al. Manipulating optical rotation in extraordinary transmission by hybrid plasmonic excitations. Appl Phys Lett 2008; 93: 021110.

24 Cheng H, Chen SQ, Yu P, Li JX, Deng L et al. Mid-infrared tunable optical polarization converter composed of asymmetric graphene nanocrosses. Opt Lett 2013; 38: 15671569.

25 Grady NK, Heyes JE, Chowdhury DR, Zeng Y, Reiten MT et al. Terahertz metamaterials for linear polarization conversion and anomalous refraction. Science 2013; 340: 13041307.

26 Shaltout A, Liu JJ, Shalaev VM, Kildishev AV. Optically active metasurface with non-chiral plasmonic nanoantennas. Nano Lett 2014; 14: 44264431.

27 Jiang SC, Xiong X, Hu YS, Jiang SW, Hu YH et al. High-efciency generation of circularly polarized light via symmetry-induced anomalous reection. Phys Rev B 2015; 91: 125421.

28 Li JX, Chen SQ, Yang HF, Li JJ, Yu P et al. Simultaneous control of light polarization and phase distributions using plasmonic metasurfaces. Adv Funct Mater 2015; 25: 704710.

29 Shitrit N, Nechayev S, Kleiner V, Hasman E. Spin-dependent plasmonics based on interfering topological defects. Nano Lett 2012; 12: 16201623.

30 Yang YM, Wang WY, Moitra P, Kravchenko II, Briggs DP et al. Dielectric meta-reectarray for broadband linear polarization conversion and optical vortex generation.

Nano Lett 2014; 14: 13941399.31 Yu P, Chen SQ, Li JX, Cheng H, Li ZC et al. Generation of vector beams with arbitrary spatial variation of phase and linear polarization using plasmonic metasurfaces.

Opt Lett 2015; 40: 32293232.32 Zhao Y, Al A. Tailoring the dispersion of plasmonic nanorods to realize broadband optical meta-wave plates. Nano Lett 2013; 13: 10861091.

33 Yu NF, Aieta F, Genevet P, Kats MA, Gaburro Z et al. A broadband, background-free quarter-wave plate based on plasmonic metasurfaces. Nano Lett 2012; 12: 63286333.

34 Zhao Y, Al A. Manipulating light polarization with ultrathin plasmonic metasurfaces. Phys Rev B 2011; 84: 205428.

35 Li T, Wang SM, Cao JX, Liu H, Zhu SN. Cavity-involved plasmonic metamaterial for optical polarization conversion. Appl Phys Lett 2010; 97: 261113.

36 Baida FI, Boutria M, Oussaid R, Labeke DV. Enhanced-transmission metamaterials as anisotropic plates. Phys Rev B 2011; 84: 035107.

37 Wu S, Zhang Z, Zhang Y, Zhang KY, Zhou L et al. Enhanced rotation of the polarization of a light beam transmitted through a silver lm with an array of perforated S-shaped holes. Phys Rev Lett 2013; 110: 207401.

38 Liu WW, Chen SQ, Li ZC, Cheng H, Yu P et al. Realization of broadband cross-polarization conversion in transmission mode in the terahertz region using a single-layer metasurface. Opt Lett 2015; 40: 31853188.

39 Johnson PB, Christy RW. Optical constants of the noble metals. Phys Rev B 1972; 6:

43704379.

40 Born M, Wolf E. Principles of Optics. Cambridge: Cambridge University Press; 1999.41 Gori F. Measuring Stokes parameters by means of a polarization grating. Opt Lett 1999; 24: 584586.42 Liu ZC, Chen SQ, Li JX, Cheng H, Li ZC et al. Fully interferometric controllable anomalous refraction efciency using cross modulation with plasmonic metasurfaces.

Opt Lett 2014; 39: 67636766.43 Roberts A, Lin L. Plasmonic quarter-wave plate. Opt Lett 2012; 37: 18201822.44 COMSOL Multiphysics Users Guide, Version 3.5. Comsol AB: Brulington, MA; 2008.45 Plum E, Fedotov VA, Zheludev NI. Optical activity in extrinsically chiral metamaterial.

Appl Phys Lett 2008; 93: 191911.46 Plum E, Liu XX, Fedotov VA, Chen Y, Tsai DP et al. Metamaterials: optical activity without chirality. Phys Rev Lett 2009; 102: 113902.47 Yannopapas V. Circular dichroism in planar nonchiral plasmonic metamaterials.

Opt Lett 2009; 34: 632634.48 Cao T, Cryan MJ. Enhancement of circular dichroism by a planar non-chiral magnetic metamaterial. J Opt 2012; 14: 085101.

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 articles Creative Commons license, unless indicated otherwise

inthecredit line;ifthematerial is not included under the Creative Commons license, users

will need to obtainpermissionfrom the license holder to reproduce the material.To view a

copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Supplementary Information for this article can be found on the Light: Science & Applications' website (http://www.nature.com/lsa).

doi:http://dx.doi.org/10.1038/lsa.2016.96

Web End =10.1038/lsa.2016.96

Light: Science & Applications