Abstract
We have achieved efficient polarization manipulations based on dipolar interferences and lattice couplings in one-dimensional cylindrical metalattices. First, we employ the scattering asymmetry factor g to quantify the directional scattering ability and find the maximum for a cylinder with effective excitations of electric and magnetic dipoles simultaneously. Further, the strong negative–g (gp = −0.38) for p–polarization and positive–g (gs = 0.68) for s–polarization are obtained within a narrow visible band using c-Si with experimental data. Inspired by the polarization-dependent phenomena, we design a metalattice-based linear polarizer considering lattice effects with an optimal particle arrangement. The metalattice performs near-perfect reflection for p–polarized waves but with zero reflection for s–polarized waves with large extinction ratios for transmission (17 dB) and reflection (24 dB). The perfect functionalities can be attributed to the near-field lattice couplings with dipolar interferences. And, we reveal that the polarization-dependent scattering coefficients, which are sensitive to the lattice period, can be largely tuned owing to lattice effects, therefore contributing to modifying far-field scattering patterns. More specifically, the proposed linear polarizers also show robust and reliable functionalities when considering lattice imperfections, the effects of system sizes, oblique incident angles, and the tunbility for different working wavelengths. The present study paves a way to stimulate many advanced practical implements based on multipolar interferences and lattice couplings.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
The recent emergence and rapid developments of light scattering by an individual meta-atom and metasurfaces have led to a wide range of advanced functionalities, such as optical cloaking [1], negative refractive index [2], beam splitters [3], holograms [4, 5], vortex masks [6, 7] and so on. Compared with metallic counterparts, dielectric materials [8], for example Si, Ge and GaAs, exhibit significantly lower losses in the visible and infrared regions with the possibilities to efficiently excite high-order Mie modes. Appropriate excitations and interferences between these modes pave a new paradigm for the design of metadevices with high efficiency and tunable directional functionalities, such as perfect reflectors and Huygen’s metasurfaces [9, 10], high-efficiency optical polarizers [11], etc.
In terms of directional scattering of an individual particle, the investigation was originally proposed by Kerker et al. in 1983 [12]. That is, under certain conditions for the permittivity and permeability of the particle, the optical response to plane-wave illumination consists of equal amplitude for electric dipoles (EDs) and magnetic dipoles (MDs). Though it has been well known for a long time, perfect forward scattering (FS) at visible [13] or near-infrared [14] regimes have not been experimentally demonstrated until recently, thanks to the excitations of strong magnetic response in nonmagnetic particles via artificial magnetism created by subwavelength structures [15]. Then the strong FS or backward scattering (BS) patterns can be well shaped based on instructive or destructive interferences between EDs and MDs. Inspired by the pioneering works [13–15], tailoring directional scattering through meta-dielectric or all-dielectric nanoparticles have attracted much attention [16–21]. While the Kerker’s conditions were initially derived for spherical particles [12], it has been demonstrated that the principles can be generalized to more complex geometries such as cylinders [20, 22], nanodisks [23] or nanopillars [24], etc. Besides, higher orders of Mie resonances can also be properly excited and tuned to meet the optimal FS condition, which has recently been achieved with a metal ring in THz frequencies depending on manipulating the amplitude and phase of the ED and electric quadrupole (EQ) [25]. Furthermore, controlling over the directional scattering can also be obtained using arrangements of elements like dimers [26], trimers or nanoparticles clusters [27] and even anisotropic nanostructures [20].
Although meta-atoms with desirable scattering directionality have been widely investigated, there is still something missing. Firstly, up to now achieving optimal BS is seldom discussed due to the difficulty in realizing zero-FS [28, 29] and low BS efficiency [18]. Actually, realizing perfect BS is of fundamental and practical importance in various novel applications, including optical cloaking [1], optical metacages [30], perfect reflectors [31], high-efficient polarizers [32] and so on. Secondly, despite of the pioneering work experimentally demonstrating the strong FS using dielectric materials, a number of later theoretical studies, on the contrast, paid much attention to fictitious materials with wavelength-independent optical constants [19, 21, 23]. Though the results may be instructive to some extent, the demonstrations by utilizing realistic materials with consideration of intrinsic absorption [33] are still relatively rare. Besides, the complex structures such as thin nanorings [25] are consequently more demanding for advanced fabrication technologies, making them difficult to fabricate. As a result, shaping the directional radiation patterns, especially for achieving strong BS using much simpler subwavlength nanoparticles with realistic materials is of great necessity and importance.
On the other hand, stimulated by the proposal of the generalized laws of reflection and refraction [34, 35], we have seen vibrant and flourishing developments in metasurfaces [36, 37] which have offered a completely new way to control light-matter interaction (including phase, amplitude and polarization of electromagnetic waves) in sub-wavelength scale [38–43]. Especially for Huygens’ metasurface based on Huygens’ sources (via the first Kerker’s condition) and Huygens’ reflectors (via the second Kerker’s condition), they offer a new way to realize full-wave control with high efficiency owing to resonant excitations and interferences of Mie modes [44–46]. For example, ideally 100% transmission efficiency and almost arbitrary spatial distributions of phase discontinuities have been theoretically and experimentally achieved in the near-infrared regime using arrays of silicon nanodisks [47]. Nevertheless, the role of lattice couplings [48] ascribed to particle arrangements in multipolar interferences [49] has not been discussed in detail, since the existing work mainly focused on the phenomenological performance or collective functionalities of the metasurfaces [43]. Studying lattice couplings is of great importance to unveil the physical mechanism behind novel optical phenomena. Specifically, the optical responses taking advantages of multipolar interferences and lattice couplings, can be very different from those of an isolated scatterer, providing an extra degree of freedom to design novel metadevices with improved efficiency. For example, by combining the effects of diffraction and multipolar interferences, it is achievable to steer light in certain directions with desired diffraction orders and unitary efficiency [50]. Besides, large-angle beam steering, perfect reflection and perfect transmission can also be realized when employing coherent lattice couplings to manipulate the excitation efficiencies of the optical modes [51]. Therefore, it is worth exploring the connections between them systematically.
Here, firstly, for an individual particle, we propose a principle to study the directional scattering of a cylinder based on the scattering asymmetry factor g. We show that strong BS for one polarization and strong FS for the other polarization can be obtained using realistic material (c-Si) with experimental data. Stimulated by the polarization-dependent phenomena, we design a one-dimensional metalattice to achieve a linear polarizer by combining the lattice effects and dipolar interferences. We find that lattice effects induced by the typical configuration are important in enhancing dipolar excitation efficiency, with the possibility to make the perfect metalattice-based polarizer come ture. Furthermore, the interplays between lattice couplings and dipolar interferences have been unequivocally unveiled to explain the reason for the perfect performance. Finally, we further investigate the performance of our proposed linear polarizers when considering the lattice perturbations, the effects of the system sizes, different incident angles and different working wavelengths as well.
2. Directional scattering of an individual meta-atom
2.1. Theory: Directional manipulation based on the scattering asymmetry factor g
We start by considering the directional far-field scattering patterns of an infinite dielectric cylinder with a normally incident plane wave from free space. The incident light can be s-polarized with electric field along the axis of the cylinder or p-polarized with magnetic field parallel to the axis of the cylinder shown in Fig. 1(a). The scattering field can be expanded into a set of cylindrical harmonics with the Mie scattering coefficients for s-polarization and p-polarization, respectively, in which m denotes the cylindrical harmonic number. With the scattering coefficients in the hands, the scattering characteristics in near field and far field can be obtained easily. Meanwhile, the scattering efficiency can also be directly calculated by [52, 53]
where k is the wave number in the background medium, r is the radius of the cylinder. More specifically, the can be simplified as due to the mirror symmetry of a cylinder. Note that the scattering coefficients of a single scatterer meet according to energy conservation. Besides, it is worth mentioning that the cylindrical harmonics can be mapped to electromagnetic multipoles: for s-polarization, corresponds to the ED, and , correspond to the MDs and magnetic quadrupoles (MQs), respectively. While for p-polarization, corresponds to the MD, , correspond to the EDs and EQs, respectively.Generally, the angular distributions of scattering patterns can be viewed by the differential scattering cross section dσs,p/dθ for s-polarization and p-polarization, respectively. For a cylindrical scatterer, it shows that [52, 53]
where σs,p is the scattering cross section for both polarizations and θ is the scatter angle. Also, in terms of Mie scattering coefficients, one can also read that where γm is the phase of the m-order electromagnetic modes. As such, by generalizing Kerker’s conditions to a cylinder, zero BS and near-zero FS can be obtained as long as the magnitudes of the ED and MD satisfy |a0|= 2|a1| and the phase difference meets Δγ = |γ0 − γ1| = 0 (the first Kerker’s condition) or Δγ = π (the second Kerker’s condition). However, above analyses appropriate to the normal FS or BS only with no consideration of other scattering directions which are also vital in scattering distributions [18].As an alternative, we employ the scattering asymmetry parameter g =< cosθ > [14] which is defined as the mean cosine of the scattering angle θ to measure the strength of the scattering directionality. The g related to the photon scattering and transport mean free path is a more general concept in scattering theory [52, 53], which is popular especially for single-particle scattering [54] or light transport in disordered media [55]. The g can be easily calculated based on the differential scattering cross section dσs,p/dθ showing [52, 53]
The large positive value of g indicates that particles scatter strongly in the forward direction, which is an usual phenomenon for Mie particles (like red blood cells). While negative–g values are rare, which was reported only recently with high-index dielectric sphere exhibiting equal electric and magnetic response with anti-phase [54]. Theoretically, the maximum or minimum value of g for a spherical scatterer are limited to ±1/2 only when the magnitudes of ED and MD are the same with in or out-of phase. Nevertheless, up to now research on positive–g or negative–g of a single cylinder is scarce. Here, we deduce proper conditions, for the first time to our knowledge, to achieve strong FS and BS of a cylinder for both s-polarization and p-polarization based on analysis of the scattering asymmetry parameter g.In analogy with analysis of a single sphere studied in [54] taking EDs and MDs into account, the gs,p can be further specified by inserting Eq. (2) into Eq. (3), showing that
in which is its complex conjugate. Consequently, it is easy to obtain the extremum values of gs,p satisfying especially when for a single cylinder. Surprisingly, the minimum is smaller than that of a sphere particle (−1/2), showing predictable strong BS behaviors, which has not been clarified up to now. Theoretically, under this particular condition, for a cylindrical scatterer, the stronger forward or backward scattering can be ideally realized with the conditions , Δγ = 0 or Δγ = π, which will be verified in the next part using a realistic material.2.2. Realization of strong FS and BS using a thin c-Si cylinder
Let us now investigate the case of a thin cylinder (r = 0.056 µm), where only the first two orders of the Mie expansions contribute to the scattering efficiency under normal incidence. Here we study the scattering properties of a realistic material (c-Si) [33] in the visible regime (0.4 µm ≤ λ ≤ 0.75 µm) whose intrinsic absorption is low with a small imaginary part of optical constants shown in Fig. 1(b).
The results of the scattering efficiency spectra for both polarizations are illustrated in Figs. 2(a) and 2(b) along with the g-value and phase difference Δγ of the first two dipolar modes. Considering effects of EDs and MDs is reasonable here since the contributions of MQs (EQs) or even higher optical modes are relatively small. Clearly, for s-polarized waves, the value of g is always positive showing usual FS features with two representative scattering patterns plotted on the right row. And it is also understandable that the maximum g in points a − d is smaller than because the phase difference Δγ is not ideally in-phase. In contrast, in the case of p-polarization, the changes of g are remarkable. The occurrence of the strong negative–g at two points A (λ = 0.41 µm, gp = −0.29) and C (λ = 0.52 µm, gp = −0.38), shows fine agreement with the aforementioned BS condition discussed in Sec. 2.1. Besides, we would like to stress that the value of the g studied in the paper is relatively more negative than that of a silicon sphere (λ = 1.55 µm, g = −0.15) in [14]. Though the intrinsic absorption of c-Si in short wavelengths is relatively large, the absorption effects can be totally neglected in the near-infrared regime. Moreover, there is a near-perfect condition with and Δγ ≈ 0 for strong FS occurring at λ = 0.68 µm, whose BS is well suppressed with gp = 0.70.
Furthermore, as shown on the right row of Figs. 2(a) and 2(b), as for s-polarization, the scattering pattern at point c shows strong FS, while BS is dominated at point C for p-polarization. Such polarization-dependent directional scattering features also provide a basis to realize some novel polarization-based optical devices, including polarizers or beam splitters. Next, we will show that by utilizing lattice effects based on an proper particle arrangement, a perfect linear polarizer can be achieved when the reflection can be eliminated with zero BS for one polarization and for the other polarization transmission can be suppressed at the same time when the incident waves could interfere destructively with the forward scattering.
3. Design of a perfect linear polarizer through the metalattice
3.1. Theory: Light propagation in 1D metalattices
The 1D infinite periodic metalattice with period d is illustrated in Fig. 3(a) to demonstrate what we propose before. A plane wave with wavevector k (in the x − y plane) incident on the nanocylinders whose radius and optical indices are consistent with those studied in Sec. 2. The incident light under incident angle φ with respect to the x axis can be s-polarization with electric field along the z direction (E0 ‖ z) and p-polarization with magnetic field parallel to z axis (H0 ‖ z). The general scattering problem has been analytically solved via multiple scattering theory with the consideration of the lattice couplings. For the j–th nanocylinder centered at Rj on the x − y plane (j = −N : N, 2N + 1 cylinders are considered here), its scattering filed can be expanded into a sum of a series of cylindrical harmonics, and the scattering coefficients for the j–th nanocylinder are related to the isolated-nanocylinder scattering coefficients through [52, 56]
in which is the coupling matrix between the m–th cylindrical harmonic of the j–th nanocylinder and the n–th cylindrical harmonic of the i–th nanocylinder. Here H(1) is the first kind of Hankel function, Rij = Ri − Rj and ϕij is the polar angle of the Rij. All sorts of couplings, including electric-electric multipolar couplings, magnetic-magnetic multipolar couplings, and cross magnetoelectric couplings between electric and magnetic multipoles, have been embedded into the coupling matrix Ωjm,in. The related to the positions of cylinder arrangements and scattering coefficients of an isolated cylinder can be calculated by . After solving the Eq. (5), all the scattering coefficients of each cylinder in the lattice can be directly obtained. It is worth noting that the scattering coefficients of a nanocylinder within the metalattice can be different from those of an individual counterpart [51] showing different arrangement-dependent scattering features due to the influence of lattice couplings. Those differences play a considerably important role in realizing functionalities of metasurfaces, such as perfect transmission, reflection and so on. However, the existing work merely focused on the collective performance of the metasurfaces, the differences of optical responses between a single particle and the particle in the lattice are not systematically clarified, not to mention the influence of lattice couplings in multipolar interferences, which is what we will discuss later.With the expansion coefficients obtained, for the proposed metalattice depicted in Fig. 3(a), the reflection and transmission of the v–th diffraction order can be expressed by [51, 56]
in which v is the diffraction order, kyv = ky + 2vπ/d with ky = |k|sinφ, and δ0v is the Kronecker delta. The and are the reflection and transmission amplitudes for both polarizations. Note that those equations are valid only when kxv is real [51, 56]. For a desirable metalattice-based linear polarizer where the s-polarized waves could totally penetrate and the p-polarized light could be perfectly reflected, its performance can be measured by extinction ratios for transmission βT and reflection βR defined as following which are in the unit of decibel (dB).3.2. Design of a perfect linear polarizer
Inspired by the aforementioned design principles, a polarization-dependent metalattice can be achieved in the visible regime (0.4 µm ≤ λ ≤ 0.75 µm) using a row of thin c-Si nanocylinders studied in Sec. 2.2. In this section, we first calculate the extinction ratios for transmission βT and reflection βR of two polarized light for 1D arrays with different period d, aiming to find an optimal lattice arrangement where the values of βT and βR are extremely high at a certain wavelength. The metalattice contains 2N + 1 = 201(N = 100) nanocylinders, which is sufficient to get reliable results. Here in this paper, we focus on the regime where there is only zeroth-order diffraction in the far field: and , for v ≥ 1. Consequently, the range of the lattice period is restricted within 0 < d ≤ 0.4 µm. Then the transmission and reflection can also be simply expressed as and . The calculated results of βT and βR for different periods and working wavelengths are illustrated together in Fig. 3(b), in which two curves denoting the highest value of βT (left) and βR (right) are clearly seen. Therefore, it is possible to choose an optimal lattice period d = 0.32 µm where the βT and βR are extremely large in the visible regime (λ ≈ 0.58 µm) simultaneously.
For the metalattice with the optimal period, the transmission and reflection spectra for both polarized waves are shown in Fig. 4(a) along with the results of βR (top) and βT (bottom) around λ = 0.58 µm shown in Fig. 4(b). It is clear that transmission can be well suppressed for p-polarization while the reflection is considerably rare for s-polarization, leading to an excellent performance for realization of a polarization-dependent metalattice. Notably, it is reasonable that due to the intrinsic absorption of the c-Si in the visible regime. Still, the extinction ratios for transmission βT and reflection βR at λ = 0.58 µm are impressively high with βT ≈ 17 dB and βR ≈ 24 dB, which are larger than most of researches studied recently [57, 58].
3.3. Analysis of dipolar interferences with lattice couplings
As discussed in Sec. 3.1, due to the lattice couplings, the optical features of the whole metalattice are distinct from the scattering of an isolated nanocylinder. To further clarify the underlying mechanism, firstly, we make a comparison of scattering coefficients between of the lattice-nanocylinder and of the individual nanocylinder. According to the Floquet theory [59] in a 1D system, because of with ky = 0 under normal incidence, it is reasonable to only consider here. Similarly, also meets when φ = 0. As illustrated in Fig. 5(a), obviously, for two polarizations at λ = 0.58 µm pointed with p and s in Fig. 4, the lattice couplings play a significant role in improving dipolar excitation efficiency. The magnitudes of the zeroth optical modes (MD for p-polarization and ED for s-polarization) are enhanced over the unity, giving considerable weights to modify scattering patterns of an isolated scatterer. Further, the corresponding far-field angular scattering patterns are illustrated in the right row. The scattering efficiencies at two cases have been dramatically enhanced (the red dot-dashed lines) while the influence of lattice couplings in phases of EDs and MDs is negligible. The directivity of two scattering patterns remain unchanged with only enhancement in scattering amplitude. Actually, it is understandable that, taking s-polarization as an example, the electric field E0 as well as ED moment is oriented perpendicular to the lattice wave propagation (in the y direction) where the lattice resonances involve [48]. Then, the lattice effects are more prominent for ED reasonably. The analysis can also be applicable for p-polarization to explain why the changes of scattering coefficients for MD is remarkable. On the other hand, by further expanding the differential scattering cross section defined in Eq. (3), one can immediately obtain
when merely considering the first two-order optical modes. It is obvious that the zeroth-order (m = 0) optical modes considered here have relatively insignificant effects on scattering angular distributions but with a considerable modification in enhancing scattering efficiency, which would be especially attractive in improving the performance of metasurfaces.Furthermore, to make the lattice effects more clear, in Fig. 6(a), we show the transmission and reflection spectra calculated by using the scattering coefficients of an individual cylinder considered in Sec. 2.2. The performance of such metalattices ignoring the effects of lattice couplings shows no desirable polarization-dependent functionality with low transmission for s–polarization and low reflection for p–polarization, further confirming the role of the coupling effects which ensures the large improvement in polarization conversion efficiency discussed in Sec. 3.2. Besides, the lattice couplings also have some influences in modifying scattering angular distributions. The comparison results of gs,p for both the single cylinder and the cylinder in the lattice are illustrated in Fig. 6(b). Though the changes of gs,p are not large consistent with the analysis mentioned above, the reduction of gp at short wavelengths is also remarkable, which is crucial in increasing the reflection for metasurfaces due to the existence of lattice effects. The representative scattering pattern pointed with e, as a inset in Fig. 6(b), shows that the BS has been further enhanced at λ = 0.5 µm where the value of gp is decreased from −0.35 to −0.52, which can potentially improve the polarization conversion efficiency.
In fact, even though the only zeroth diffraction is included in the far field, lattice couplings with dipolar interferences can still be affected due to the existence of high-order diffractive evanescent field. In the near field, as for the first diffraction order (v = 1), and the lattice wave vector becomes imaginary. It is possible that the evanescent filed of high-order diffractions can interplay with dipolar modes within a single meta-atom under an appropriate lattice period, which ensures the high-efficient couplings. In Fig. 7, we illustrate the electric field intensity |E|2 with three lattice periods at λ = 0.58 µm to show the difference of the lattice effects in resonance features. When the period is small with d = 0.2 µm shown in Figs. 7(a)(1) and 7(b)(1), although the lattice effects may appear, the field enhancement is relatively weak or mainly occurs in free space (air) [60], showing weak effects in enhancing scatterer resonances. Similarly, in the case of large period (d = 0.4 µm), the role of lattice effects is decreased and the scattering profile is more akin to the single scattering shown in Fig. 7(a)(3). While for the optimal lattice arrangement (d = 0.32 µm) calculated in the last part, a large field enhancement can be observed in the vicinity of the cylinders with the brightest color map illustrated in the middle row of Fig. 7. These changes will be further verified in Sec. 4.1. Actually, there is a trade-off between the decay of high-order diffraction modes and lattice-coupling strength in our systems, thus leading to an proper arrangement with desirable performance. Consequently, we can conclude that the scattering coefficients can be significantly enhanced due to the near-field couplings between dipolar interferences and lattice-induced high-order diffractive evanescent field under a proper lattice period.
Therefore, based on the design principles and detailed analysis in this section, the lattice effects on dipolar interferences have been distinguished unequivocally comparing with the single-particle scattering. The lattice couplings play a vital role not only in improving the efficiency of the metalattice but also in directional modification of scattering patterns within certain wavelengths, which suggests a new way to design and further manipulate functionalities of the metasurfaces. In the next section, we would like to show that the proposed polarizer exhibits robust and excellent performance when considering practical implementation issues.
4. Discussion on practical implementation
Above analysis indicates that the lattice effects in dipolar interferences are of great importance and the calculated results about perfect transmission, reflection and large extinction ratios of the metalattice further confirm the reliability of our design principles proposed in this paper. In this section, we further discuss some inevitable problems in practical implementation of the designed metalattice, especially for the tolerance of lattice imperfections, the effects of the system size, the dependence on incident angles and the applicability for different working wavelengths as well.
4.1. Tolerance of lattice imperfections
In practical fabrication, it is inevitable to suffer some lattice imperfections including the nonuniformity in diameters of the nanocylinders, the irregular shapes of the cross sections, small but non-negligible disorder in metaatom positions and so on [58]. Here in this part we mainly focus on the disordered perturbations in cylinder arrangements and further indicate the central role of perfect lattice effects. The perturbation Δd is introduced to measure the position deviations from the perfect lattice arrangements. The range of the perturbation Δd is chosen to be , aiming to avoid the overlap between adjacent nanocylinders. Then the disordered degree can be defined as χ = |Δd|/|Δd|max ranging with 0 ≤ χ ≤ 1.
In Fig. 8, we present how the functionalities of the proposed polarizers (λ = 0.58 µm, period d′ = d + Δd) are influenced by lattice disorder, in which all the results are calculated by assemble averaging 50 different configurations. It can be seen in Fig. 8(b) that both βT and βR experience apparent decreases as the disordered degree increases. Even though the performance of the polarizer with lattice imperfections are damaged compared with the perfect periodic counterpart, the functionality can still be preserved with βT > 10 dB and βR > 20 dB for χ ≤ 0.6. When the disordered degree increases to χ = 1, the extinction ratios drop to βT ≈ 8 dB and βR ≈ 17 dB which are still acceptable in practical implement. Notably, in Fig. 8(a), we also show the absorption for two polarizations with . The disorder imperfections introduced here play important roles in reducing transmission for s–polarization and weakening reflection for p–polarization but few significant change in and , leading to the considerable absorption enhancement as the disordered degree increases [61]. Besides, the absorption enhancement for p polarization is larger than that of s polarization, since for p-polarized waves electric fields are in the x − y plane and multiple scattering can be further strengthened ascribed to the occurrence of disorder in lattice.
Meanwhile, we further investigate how the strength of lattice couplings responses to the lattice imperfections. According to the analysis in Sec. 3.3, results indicate that lattice effects play a dominating role in improving scattering efficiency of ED for s-polarization and MD for p-polarization. Here we define a coupling strength ηs,p to measure the strength of lattice couplings with dipoles showing
In Fig. 8(c), we show the changes of ηs,p at λ = 0.58 µm with different degrees of lattice imperfections as a function of the lattice period considered in Sec. 3.2. Obviously, in the case of the perfect meta-atom arrangement (χ = 0), the strength of lattice couplings is considerably large with ηs,p > 1.5 for both polarizations, indicating the vital role of lattice effects. As the disordered degree increases, the perfect lattice is destroyed, leading to an obvious decreasing of coupling strength. On the other hand, the results of ηs,p suggest that the optimal lattice period occurs around 0.32 µm for both polarizations simultaneously, which is consistent with the results in Sec. 3.3. However, note that, by combining the advantages of lattice couplings and dipolar interferences, the designed metalattice still show feasible performance even though the disordered perturbations occur.4.2. Effect of system size
Above numerical calculations are performed with a metalattice consisting of 2N + 1 = 201 (N = 100) cylinders. Since the existence of edge effects in finite systems, it is necessary and as well as of practical significance to further investigate the dependence on system size, evaluating whether a larger or smaller metalattice can still perform well with the same lattice period.
In Fig. 9, we show how the performance parameters of the proposed linear polarizers vary with the sizes of the metalattices, where the calculation is conducted to the largest system with 201 nanocylinders. Obviously, the transmission , reflection and the performance of extinction ratios for transmission βT and reflection βR are all convergent in the end. It can be observed that for the systems containing over 41 (N = 20) nanocylinders, the performance (βT and βR) of the metalattices can be well preserved, indicating that the designed linear polarizer can still operate for large-size systems. While for small cases, the influences of size effects in βT and βR are significant with a large fluctuation especially for extremely small systems. In the case of the N = 10 shown in Fig. 9, the extinction ratios for transmission βT ≈ 17 dB remains relatively stable, while the βR witnesses a sharp increase with 40 dB where the edges effects are more substantial and the Floquet theory may be not valid. Results suggest that careful modification of the lattice period should be considered when metalattices operating with small sizes. Even so, as for all the cases calculated here, the extinction ratios for transmission βT and reflection βR can still be ideally higher with βR > 18 dB and βT > 15 dB, ensuring the excellent and reliable performance in practical implement.
4.3. Acceptance of incidence angles
In this part, we study the effects of oblique incidence on performance of the designed metalattices possessing excellent polarization dependent features. Since we merely focus on the diffractionless regime, the range of incident angles of interest is restricted within where only zeroth order diffraction should be considered. When φ is larger than 54°, higher multipoles may be excited and those can interference with each other as well as with lattice couplings in the far field, leading to the generation of higher diffraction orders, which is not taken into account in this paper.
The results of the extinction ratios for transmission βT and reflection βR at λ = 0.58 µm varying with the incident angles are illustrated in Fig. 10(b) along with the reflection and transmission spectra depicted in Fig. 10(a). It is clear that under oblique incidence, the performance of the metalattice is unavoidably destroyed to some extent. When the incident angle is smaller than around 10°, the optical response of the polarizer is reliable with both βT and βR larger than 10 dB. The acceptance of the incident angles for the 1D metalattice can be attributed to the high refractive index contrast between nanocylinders (c-Si) and background (air). On the other hand, the nanocylinders are subwavelength in diameter, which also reduces the shadow effects from adjacent metaatoms [62]. But the significant decrease for large oblique angles cannot be ignored in Fig. 10(b), suggesting that the incident conditions should be well controlled within the acceptance range, or lattice period and radius of nanocylinders should be further modified to well preserve the performance of the designed polarizer.
4.4. Realization of polarizers with different working wavelengths
According to the design principles discussed in this paper, reliable functionalities of the linear polarizers can be achieved. Since the perfect directional conditions are considered here, the bandwidth of the metalattice-based polarizer is not broad seen from Fig. 4(b). Nevertheless, the working wavelengths can be tuned by properly changing the lattice period and the geometric parameters of nanocylinders. Next, we will show for different target working wavelengths the linear polarizers can be also achieved in the visible range. The obtained desirable extinction ratios for transmission βT and reflection βR as well as corresponding the optimal lattice period d and cylinder’s radius r are illustrated in Fig. 11.
In terms of different target wavelengths, as shown in Fig. 11(a), it is possible to find appropriate parameters with an optimal period d and a radius r of cylinders based on the numerical analysis in Sec. 3.2. The extinction ratios for λ > 0.55 µm considered here are considerably large with βT, βR > 15 dB, which are desirable in practical applications. Besides, it can be seen that as the working wavelength increases, the optimal lattice period and the radius of cylinders also increase linearly, which agrees well with the scale invariance rule of Maxwell equations since the optical constants within 0.55 µm < λ < 0.75 µm are almost similar shown in Fig. 1(b). Note that the results when λ < 0.55 µm are not illustrated here because the intrinsic absorption of c-Si in short wavelengths is too large, thus the polarization characteristics are destroyed inevitably. In contrast, the absorption loss among long wavelengths is negligible and the outstanding performance can be preserved, which still offers a promising way to design precise optical devices.
5. Conclusion
In this work, we propose the design principles to achieve excellent directional polarization manipulations of an individual meta-atom and metelattices by taking the dipolar interferences and lattice couplings into consideration. Firstly, we employ the scattering asymmetry factor g to investigate the scattering directionality of an individual cylindrical particle in the visible regime and find the extremum value of the for a cylinder. The strong forward (gs = 0.64; gp = 0.70) and backward (gp = (0.38) scattering can be observed in two different polarized waves simultaneously owing to the efficient excitation of EDs and MDs.
By combining lattice effects and dipolar interferences, we propose a metalattice-based linear polarizer at the visible wavelength stimulated by the aforementioned polarization-dependent phenomena. After properly designing the lattice period d = 0.32 µm, the transmission light through the designed metalattice can be perfectly vanished for p-polarization and for s-polarization reflection is also eliminated, showing excellent polarization-selective performance with the large extinction ratios (βT ≈ 17 dB, βR ≈ 24 dB). Furthermore, we have theoretically discussed the underlying mechanism for its feasible functionality. Analyses show that the efficiency of dipolar excitations is enhanced significantly ascribed to the interplays between the dipolar modes and high-order diffractive evanescent field with an appropriate lattice arrangement, leading to a considerable modifications of scattering patterns of an isolated scatterer. Besides, we also reveal that for 1D cylindrical systems, the connections between dipolar interferences and lattice couplings show obvious polarization selectivity. As for s-polarized waves, the lattice couplings mainly affect the profile of ED, while such influence in MD is prominent for p-polarization. On the other hand, the proposed linear polarizer shows stable and reliable performance in practical implement. The large extinction ratios for transmission βT and reflection βR can be preserved despite of the existence of lattice imperfections, while the absorption is enhanced as the disordered degree increases, which also indicates that the optimal lattice couplings emerge in the case of perfect lattice arrangement (χ = 0). In addition, the insensitivity to the system sizes (2N + 1 > 41) and the acceptance of the incident angles (φ < 10°) of our metalattice are also attractive and desirable in many applications. Furthermore, we also show that polarizers with different working wavelengths can be also obtained by tuning the lattice period and the radius of cylinders. The theory and designing principles in this paper provide a promising platform to investigate the effects of dipolar interferences and lattice couplings, which can also potentially stimulate more novel optical phenomena and applications.
Funding
The National Natural Science Foundation of China (51636004, 51476097); Shanghai Key Fundamental Research Grant (16JC1403200); The Foundation for Innovative Research Groups of the National Natural Science Foundation of China (51521004).
References and links
1. Z. J. Wong, X. Ni, M. Mrejen, Y. Wang, and X. Zhang, “An ultrathin invisibility skin cloak for visible light,” Science 349(6254), 1310–1314 (2015). [CrossRef] [PubMed]
2. B. A. Slovick, “Negative refractive index induced by percolation in disordered metamaterials,” Phys. Rev. B 95, 094202 (2016). [CrossRef]
3. K. B. Crozier, M. Khorasaninejad, and W. Zhu, “Efficient polarization beam splitter pixels based on a dielectric metasurface,” Optica 2(4), 376–382 (2015). [CrossRef]
4. X. Ni, A. V. Kildishev, and V. M. Shalaev, “Metasurface holograms for visible light,” Nat. Commun. 4, 2807 (2013). [CrossRef]
5. W. Zhao, H. Jiang, B. Liu, J. Song, Y. Jiang, C. Tang, and J. Li, “Dielectric Huygens’ Metasurface for High-Efficiency Hologram Operating in Transmission Mode,” Sci. Rep. 6, 30613 (2016). [CrossRef]
6. A. Zhan, S. Colburn, R. Trivedi, T. K. Fryett, C. M. Dodson, and A. Majumdar, “Low contrast dielectric metasurface optics,” in Lasers and Electro-Optics, (2016), paper JW2A.14.
7. Z. Wang, Y. Yan, A. Arbabi, G. Xie, C. Liu, Z. Zhao, Y. Ren, L. Li, N. Ahmed, and A. J. Willner, “Orbital angular momentum beams generated by passive dielectric phase masks and their performance in a communication link,” Opt. Lett. 42(14), 2746–2749 (2017). [CrossRef] [PubMed]
8. S. Jahani and Z. Jacob, “All-dielectric metamaterials,” Nat. Nanotechnol. 11(1), 23–36 (2016). [CrossRef] [PubMed]
9. M. Decker, I. Staude, M. Falkner, J. Dominguez, D. N. Neshev, I. Brener, T. Pertsch, and Y. S. Kivshar, “High–Efficiency Dielectric Huygens’ Surfaces,” Adv. Opt. Mater. 3(6), 813–820 (2015). [CrossRef]
10. S. Liu, A. Vaskin, S. Campione, O. Wolf, M. B. Sinclair, J. Reno, G. A. Keeler, I. Staude, and I. Brener, “Huygens’ Metasurfaces Enabled by Magnetic Dipole Resonance Tuning in Split Dielectric Nanoresonators,” Nano Lett. 17(7), 4297–4303 (2017). [CrossRef] [PubMed]
11. S. Kruk, B. Hopkins, I. I. Kravchenko, A. Miroshnichenko, D. N. Neshev, and Y. S. Kivshar, “Invited Article: Broadband highly efficient dielectric metadevices for polarization control,” APL Photonics 1, 030801 (2016). [CrossRef]
12. M. Kerker, D. S. Wang, and C. L. Giles, “Electromagnetic scattering by magnetic spheres,” J. Opt. Soc. Am. 73, 765–767 (1983). [CrossRef]
13. Y. H. Fu, A. I. Kuznetsov, A. E. Miroshnichenko, Y. F. Yu, and B. Luk’Yanchuk, “Directional visible light scattering by silicon nanoparticles,” Nat. Commun. 4, 1527 (2013). [CrossRef] [PubMed]
14. M. Nietovesperinas, R. Gomezmedina, and J. J. Saenz, “Angle-suppressed scattering and optical forces on submicrometer dielectric particles,” J. Opt. Soc. Am. A 28(1), 54–60 (2011). [CrossRef]
15. J. M. Geffrin, B. García-Cámara, R. Gómez-Medina, P. Albella, L. S. Froufe-Pérez, C. Eyraud, A. Litman, R. Vaillon, F. González, and M. Nieto-Vesperinas, “Magnetic and electric coherence in forward- and back-scattered electromagnetic waves by a single dielectric subwavelength sphere,” Nat. Commun. 3, 1171 (2012). [CrossRef] [PubMed]
16. B. Lei, H. Hu, H. Ma, J. Zhang, W. Liu, and W. Xie, “Ultra-directional forward scattering by individual core-shell nanoparticles,” Opt. Express 22(13), 16178–16187 (2014). [CrossRef] [PubMed]
17. A. Pors, S. K. Andersen, and S. I. Bozhevolnyi, “Unidirectional scattering by nanoparticles near substrates: generalized Kerker conditions,” Opt. Express 23(22), 28808–28828 (2015). [CrossRef] [PubMed]
18. P. R. Wiecha, A. Cuche, A. Arbouet, C. Girard, G. C. D. Francs, A. Lecestre, G. Larrieu, F. Fournel, V. Larrey, and T. Baron, “Strongly Directional Scattering from Dielectric Nanowires,” ACS Photonics 4, 2036–2046 (2017). [CrossRef]
19. L. Ge, L. Liu, S. Dai, J. Chai, Q. Song, H. Xiang, and D. Han, “Unidirectional scattering induced by the toroidal dipolar excitation in the system of plasmonic nanoparticles,” Opt. Express 25(10), 10853–10862 (2017). [CrossRef] [PubMed]
20. W. Liu, “Superscattering pattern shaping for radially anisotropic nanowires,” Phys. Rev. A 96, 023854 (2017). [CrossRef]
21. W. Liu, “Ultra-directional super-scattering of homogenous spherical particles with radial anisotropy,” Opt. Express 23(11), 5489–5497 (2015).
22. W. Liu, A. E. Miroshnichenko, R. F. Oulton, D. N. Neshev, O. Hess, and Y. S. Kivshar, “Scattering of core-shell nanowires with the interference of electric and magnetic resonances,” Opt. Lett. 38(14), 2621–2624 (2013). [CrossRef] [PubMed]
23. I. Staude, A. E. Miroshnichenko, M. Decker, N. T. Fofang, S. Liu, E. Gonzales, J. Dominguez, T. S. Luk, D. N. Neshev, and I. Brener, “Tailoring directional scattering through magnetic and electric resonances in subwavelength silicon nanodisks,” ACS Nano 7(9), 7824–7832 (2013). [CrossRef] [PubMed]
24. J. F. Algorri, B. García-Cámara, A. Cuadrado, J. M. Sánchez-Pena, and R. Vergaz, “Selective Dielectric Metasurfaces Based on Directional Conditions of Silicon Nanopillars,” Nanomaterials 7(7), 177 (2017). [CrossRef]
25. R. Alaee, R. Filter, D. Lehr, F. Lederer, and C. Rockstuhl, “A generalized Kerker condition for highly directive nanoantennas,” Opt. Lett. 40(11), 2645–2648 (2015). [CrossRef] [PubMed]
26. J. Yan, P. Liu, Z. Lin, H. Wang, H. Chen, C. Wang, and G. Yang, “Directional Fano Resonance in a Silicon Nanosphere Dimer,” ACS Nano 9(3), 2968–2980 (2015). [CrossRef] [PubMed]
27. R. Dezert, P. Richetti, and A. Baron, “Isotropic Huygens dipoles and multipoles with colloidal particles,” Phys. Rev. B 96, 180201 (2017). [CrossRef]
28. N. Engheta, “How does zero forward-scattering in magnetodielectric nanoparticles comply with the optical theorem?” J. Nanophotonics 4, 041590 (2010). [CrossRef]
29. J. Y. Lee, A. E. Miroshnichenko, and R.-K. Lee, “Reexamination of Kerker’s conditions by means of the phase diagram,” Phys. Rev. A 96, 043846 (2017). [CrossRef]
30. M. Ali, A. E. Miroshnichenko, I. V. Shadrivov, and Y. S. Kivshar, “Optical Metacages,” Phys. Rev. Lett. 115, 215501 (2015). [CrossRef]
31. A. Epstein and G. V. Eleftheriades, “Synthesis of Passive Lossless Metasurfaces Using Auxiliary Fields for Reflectionless Beam Splitting and Perfect Reflection,” Phys. Rev. Lett. 117, 256103 (2016). [CrossRef] [PubMed]
32. M. H. Alizadeh and B. M. Reinhard, “Highly efficient and broadband optical polarizers based on dielectric nanowires,” Opt. Express 25(19), 22897–22904 (2017). [CrossRef] [PubMed]
33. E. D. Palik, Handbook of optical constants of solids (Academic, 1985).
34. N. Yu, P. Genevet, M. A. Kats, F. Aieta, J. P. Tetienne, F. Capasso, and Z. Gaburro, “Light Propagation with Phase Discontinuities Reflection and Refraction,” Science 334(6054), 333–337 (2011). [CrossRef] [PubMed]
35. A. V. Kildishev and V. M. Shalaev, “Planar Photonics with Metasurfaces,” Science 339(6125), 12889–1289 (2013). [CrossRef]
36. D. Lin, P. Fan, E. Hasman, and M. L. Brongersma, “Dielectric gradient metasurface optical elements,” Science 345(6194), 298–302 (2016). [CrossRef]
37. V. S. Asadchy, A. Wickberg, A. Díazrubio, and M. Wegener, “Eliminating Scattering Loss in Anomalously Reflecting Optical Metasurfaces,” ACS Photonics 4(5), 1264–1270 (2017). [CrossRef]
38. A. Dĺazrubio, V. S. Asadchy, A. Elsakka, and S. A. Tretyakov, “From the generalized reflection law to the realization of perfect anomalous reflectors,” Sci. Adv. 3, e1602714 (2017). [CrossRef]
39. A. Arbabi, H. Yu, M. Bagheri, and A. Faraon, “Dielectric metasurfaces for complete control of phase and polarization with subwavelength spatial resolution and high transmission,” Nat. Nanotechnol. 10(11), 937–943 (2015). [CrossRef] [PubMed]
40. S. S. Kruk, R. Camachomorales, L. Xu, M. Rahmani, D. A. Smirnova, L. Wang, H. H. Tan, C. Jagadish, D. N. Neshev, and Y. S. Kivshar, “Nonlinear Optical Magnetism Revealed by Second-Harmonic Generation in Nanoantennas,” Nano Lett. 17, 3914–3918 (2017). [CrossRef] [PubMed]
41. Y. Liu, Y. Ke, H. Luo, and S. Wen, “Photonic spin Hall effect in metasurfaces: a brief review,” Nanophotonics 6(1), 51–70 (2016). [CrossRef]
42. Z. Wei, Y. Cao, X. Su, Z. Gong, Y. Long, and H. Li, “Highly efficient beam steering with a transparent metasurface,” Opt. Express 21(9), 10739–10745 (2013). [CrossRef] [PubMed]
43. H. T. Chen, A. J. Taylor, and N. Yu, “A review of metasurfaces: physics and applications,” Rep. Prog. Phys. 79, 076401 (2016). [CrossRef] [PubMed]
44. S. S. Kruk and Y. S. Kivshar, “Functional meta-optics and nanophotonics govern by Mie resonances,” ACS Photonics 4(11), 2638–3649 (2017). [CrossRef]
45. S. Liu, A. Vaskin, S. Campione, O. Wolf, M. B. Sinclair, J. Reno, G. A. Keeler, I. Staude, and I. Brener, “Huygens’ Metasurfaces Enabled by Magnetic Dipole Resonance Tuning in Split Dielectric Nanoresonators,” Nano Lett. 17(7), 4297–4303 (2017). [CrossRef] [PubMed]
46. A. I. Kuznetsov, A. E. Miroshnichenko, M. L. Brongersma, Y. S. Kivshar, and B. Luk’Yanchuk, “Optically resonant dielectric nanostructures,” Science 35(6314), aag2472 (2016). [CrossRef]
47. M. Decker, I. Staude, M. Falkner, J. Dominguez, D. N. Neshev, I. Brener, T. Pertsch, and Y. S. Kivshar, “High–Efficiency Dielectric Huygens’ Surfaces,” Adv. Opt. Mater. 3(6), 813–820 (2015). [CrossRef]
48. V. E. Babicheva and A. B. Evlyukhin, “Resonant Lattice Kerker Effect in Metasurfaces With Electric and Magnetic Optical Responses,” Laser Photon. Rev. , 11, 1700132 (2017). [CrossRef]
49. W. Liu and Y. S. Kivshar, “Multipolar interference effects in nanophotonics,” Phil. R. Soc. Lond. A 375, 20160317 (2016). [CrossRef]
50. Y. Ra’Di, D. L. Sounas, and A. Alu, “Metagratings: Beyond the Limits of Graded Metasurfaces for Wave Front Control,” Phys. Rev. Lett. 119, 067404 (2017). [CrossRef]
51. W. Liu and A. E. Miroshnichenko, “Beam Steering with Dielectric Metalattices,” ACS Photonic, https://arxiv.org/abs/1710.03380 (2017).
52. H.C. van de Hulst, Light Scattering by small particles (Dover, 1981).
53. C.F. Bohren and D.R. Huffman, Absorption and Scattering of Light by Small Particles (JohnWiley and Sons, 1998). [CrossRef]
54. R. Gomezmedina, L. Froufeperez, M. Yepez, F. Scheffold, M. Nietovesperinas, and J. J. Saenz, “Negative scattering asymmetry parameter for dipolar particles: Unusual reduction of the transport mean free path and radiation pressure,” Phys. Rev. A 85, 035802 (2012). [CrossRef]
55. F. A. Pinheiro, “New effects in light scattering in disordered media and coherent backscattering cone: systems of magnetic particles,” Phys. Rev. Lett. 84(7), 1435–1438 (2000). [CrossRef] [PubMed]
57. W. Liu, “Generalized Magnetic Mirrors,” Phys. Rev. Lett. 119, 123902 (2017). [CrossRef]
58. B. Shen, P. Wang, R. Polson, and R. Menon, “An integrated-nanophotonics polarization beamsplitter with 2.4×2.4µm2 footprint,” Nat. Photonics 9(6), 378–382 (2015). [CrossRef]
59. B. X. Wang, C. Y. Zhao, Y. H. Kan, and T. C. Huang, “Design of metasurface polarizers based on two-dimensional cold atomic arrays,” Opt. Express 25(16), 18760–18773 (2017). [CrossRef] [PubMed]
60. E. N. Bulgakov and A. F. Sadreev, “Bloch bound states in the radiation continuum in a periodic array of dielectric rods,” Phys. Rev. A 90, 053801 (2014). [CrossRef]
61. W. Liu and A. E. Miroshnichenko, “Scattering Invisibility With Free–Space Field Enhancement of All–Dielectric Nanoparticles,” Laser Photon. Rev. 11, 1700103 (2017). [CrossRef]
62. A. Mafi, “Transverse Anderson localization of light: a tutorial,” Adv. Opt. Photon. 7(3), 459–515 (2015). [CrossRef]
63. D. Lin, M. Melli, E. Poliakov, P. S. Hilaire, S. Dhuey, C. Peroz, S. Cabrini, M. Brongersma, and M. Klug, “Optical metasurfaces for high angle steering at visible wavelengths,” Sci. Rep. 7, 2286 (2017). [CrossRef] [PubMed]