Abstract
One-dimensional photonic crystals (PCs), when operating near the band edge in the dispersion diagram, inherently possess nearly polarization-independent angular selectivity—an angular transmission window around the normal direction with reflection for other angles. However, the incident light is mostly reflected at the PC–air interface due to large impedance mismatch. We show that the reflection may be sufficiently suppressed by utilizing a specially designed antireflection structure consisting of a PC having a different pitch from that of the host PC. The underlying mechanism is that the interfaces of the antireflection PC with the host PC and the air structure are selected such that the transverse impedance has a real value, which is positioned at the center of the thickness of a material film. Moreover, our structure provides a high-throughput wide angular transmission window, including the normal direction in both and polarizations. We develop an analytical model that captures the angular selectivity observed in numerical results.
© 2016 Optical Society of America
One-dimensional (1D) photonic crystals (PCs) that consist of an alternative stack of different material films provide an opportunity to manipulate the dispersion relation [1] and have found numerous optical and thermal applications [2–7]. Particularly, angular selectivity is among the important functionalities of PCs and has been extensively investigated [8,9]. In Ref. [10], an angular filter for polarization has been experimentally realized in the full visible spectrum by utilizing Brewster modes, providing an angular transmission window of 8° in an oblique direction of 55°. In Refs. [11,12], a transmission window for polarization has been set to the normal direction through the use of anisotropic layers. In Ref. [13], an ultrabroadband plasmonic Brewster transmission for polarization has been presented by utilizing a metallic grating. In general, a polarization-independent angular transmission window in the normal direction is desirable for many application scenarios, such as optical sensor systems [14] and communication systems [15]. However, to the best of our knowledge, there has been no work on a polarization-independent angular filter having a high-throughput transmission window in the normal direction.
PCs exhibit polarization-independent omnidirectional reflection when operating in the bandgap of the dispersion diagram [16]. Using such bandgap operation, a defect layer in PCs allows light transmission at a specific frequency as a Fabry–Perot resonator [1,17–21], and the angular transmission window is very narrow due to the intrinsic nature of resonance. PCs inherently possess a nearly polarization-independent angular transmission window that includes the normal direction when operating near the band edge in the dispersion diagram. However, the incident light is mostly reflected back at the PC–air interface due to large impedance mismatch. The wave impedance of PCs is dependent on local observation points, and, thus, the antireflection in the extremely small wavevector regime is challenging.
In this Letter, we show analytically and numerically that the reflection of the incident light is sufficiently suppressed in the extremely small wavevector regime by utilizing an antireflection PC having a different pitch from that of the host PC, resulting in a high-throughput angular transmission filter. The underlying mechanism is that the antireflection PC needs to have the interfaces with the host PC and the air structure such that the transverse impedance at each interface between two semi-infinite media has a real value, which allows us to design the antireflection PC by using the general methodology of impedance matching. In Ref. [22], we presented an antireflection structure for a two-dimensional PC in the extremely small wavevector regime. However, it showed only a sharp angular and frequency response.
We start by considering the dispersion diagram of a PC, as shown in Figs. 1(a) and 1(b), obtained by using the plane wave expansion (PWE) method [23]. The PWE method is widely used for analyses of PCs in which Maxwell’s equations are exactly solved by expanding the electromagnetic fields with a plane-wave basis set, and, thus, the wavevectors and electromagnetic fields are accurately evaluated. Two material films are alternatively and infinitely stacked in the PC [inset of Fig. 1(a)]. We assume and for two materials with refractive indices of and , respectively. The thicknesses of the two materials are selected at and such that the PC has an angular transmission window of [oblique black dashed lines closer to the vertical axis in Fig. 1(a)] for and polarizations at the design frequency of (horizontal pink solid line). We note that extended modes that are allowed to propagate exist within angles of less than (shaded region) for both polarizations at . In other angles at , no modes are allowed to propagate (white region) except for large angles in polarization. Therefore, we expect that the PC may work as an angular filter having a nearly polarization-independent transmission window of around the normal direction, and a reflection in other angles. The effect of the leakage of electromagnetic waves in such large angles () in polarization may be reduced by optimizing the refractive indices and the thicknesses of materials.
Figure 1(c) shows the configuration of a PC having the angular selectivity function. Two material films with refractive indices of and are alternatively stacked. The films are numbered from to , where is the total number of layers. From the viewpoint of the underlying mechanism, the heterostructure consists of the host PC ( to ) and the antireflection PCs at both sides ( to and to ), where each of two layers and has the interface between the antireflection PC and the host PC positioned at the center of the thickness of the layer [indicated by red dotted lines in Fig. 1(c)] and is counted for the layer numbers of both the host and antireflection PCs. From the geometrical point of view, material has the same thickness in both the host PC and the antireflection PC except for the first layer and the last layer . The two layers at both ends have the half-thickness such that the antireflection PC has the same impedance at the interface with the air structure and at the interface with the host PC [indicated by red dotted lines in Fig. 1(c)]. Material , however, has two different thicknesses in the entire structure: for the host PC and for the antireflection PC.
The monochromatic electromagnetic waves with polarization or polarization are impinged on the entire PC from the air, where the electric field or the magnetic field is perpendicular to the incident plane, respectively. The Fresnel reflection and transmission coefficients of the electric field for polarization and the magnetic field for polarization at the interface between layers and are given by [24]
where and are the (normal to the surface of the layers) and the (parallel to the surface of the layers) components of the wavevector, is the free-space wavenumber, and is the refractive index of the th layer. The air regions are labeled by and in Fig. 1(c). The subscripts and represent the polarization. The reflection and transmission coefficients , of the entire structure of PC in air can be calculated by using Eqs. (1a)–(1d) and the transfer matrix method [25], which is widely used for analyses of multilayer structures. We assume the lossless model, , where for polarization and for polarization, and plot the power transmission coefficient for numerical results throughout the Letter.Figure 1(d) shows the two-dimensional map of the transmission coefficients of the entire structure of Fig. 1(c) with variations of the incident angle and the operating frequency. Due to the properly designed antireflection structure, high transmission (bright-color region) is obtained for both polarizations in the propagation-allowed region that is characterized by the dispersion diagram (shaded regions) in Figs. 1(a) and 1(b). It should be emphasized that the reflection is sufficiently suppressed in a wide range of frequency in the propagation-allowed region; our structure provides not only a transmission window of at but also wide angular transmission window at high frequency. When for polarization around in Fig. 1(d), the incident light is leaked out, as observed from the dispersion diagram.
We plot the transmission coefficient as a function of the incident angle at in Fig. 1(e). We clearly observe that our structure has efficient angular selectivity; light passes through around the normal direction and is reflected over other angles. The half-power angular width is 11° for polarization and 13° for polarization, respectively. In addition, at a frequency of 515 THz in Fig. 1(f), a wide transmission window appears: 57° for - polarization and 69° for polarization. Interestingly, sufficient antireflection is maintained when the frequency goes up. This wideband characteristic of the sufficient antireflection comes from the fact that the antireflection structure consists of a PC; the antireflection structure has a trend in the frequency response similar to that of the host medium.
We next elucidate how sufficient antireflection, as discussed above, is obtained. The underlying mechanism can be understood in the impedance behavior. Consider the impedance behavior at in Fig. 2. The transverse impedance for or polarization, respectively, is defined as
We assume infinite PCs for impedance analysis throughout this Letter and thus obtain each transverse impedance in Eqs. (2a) and (2b) by the PWE method. Figure 2(a) shows the locally dependent transverse impedance of the host PC at normal incidence. The transverse impedance is normalized with respect to the free-space intrinsic impedance . We point out that the imaginary part of the impedance is zero at the center of the thickness of each film [indicated by red dotted lines and in Fig. 2(a)]. This allows us to obtain sufficient antireflection by using the general methodology of impedance matching. However, at interface , which is a regular interface between layers, the impedance has a non-zero value in the imaginary part, and, thus, antireflection is in general difficult. In our case, we have selected interface since the real part of the impedance at interface is smaller than that at interface .
Figure 2(b) shows the angular dependence of the transverse impedance (blue solid line) of the host PC at interface . The real part of the transverse impedance is presented since the imaginary part is zero even for oblique incidence. The impedance has a slowly varying characteristic within for both polarizations. In addition, we calculate the transverse impedance of the antireflection PC that has a different thickness of for films with the same thickness of for films and plot in the same figure, where for polarization and for polarization. We verify that the general antireflection condition [25] is satisfied within in Fig. 2(b). The length of the antireflection PC has been selected at a quarter-guided-wavelength (number of layers: ) at according to the general antireflection condition [25]. Therefore, the reflection at the interface of the entire structure has been sufficiently suppressed, as shown in the high transmission coefficient of Fig. 1(e).
Based on the underlying mechanism above, we now develop an analytical model, as shown in Fig. 3. The heterostructure consists of three layers; the host PC having the antireflection PCs at both sides in air. The reflection and transmission coefficients at the interfaces of the analytical model of Fig. 3 are given by
where the host PC, the antireflection PC, and the air structure are labeled , , and , respectively. The transverse impedances and wavevectors are obtained in each polarization by the PWE method, and then the reflection and transmission coefficients at each interface are obtained using Eqs. (3a)–(3d). In addition, using the transfer matrix method, we calculate the reflection and transmission coefficients of the entire structure and plot the transmission coefficients in Figs. 1(e) and 1(f). The analytical results (pink dots) agree excellently with numerical results (blue solid lines). Therefore, we verify the underlying mechanism of our structure of Fig. 1(c). When for polarization at [Fig. 1(e)], the analytical result has two peaks (pink dots), while the numerical result has four peaks (blue solid line). This discrepancy is due to the assumption of the infinite PCs in our analysis; the angular range, where the transmission peaks occur in the finite PC of Fig. 1(c) in the numerical result, falls in the bandgap in the infinite PC in the analytical result.We have used a number of layers in the entire structure. The number of layers can be reduced; only the number of layers in the host PC is reduced, while that in the antireflection PCs () is maintained. Reducing the number of layers gives gradual slope in the angular response.
In conclusion, we have explored a 1D PC that operates in the extremely small wavevector regime and, therefore, has an angular transmission window around the normal direction. Based on the analysis of the locally dependent impedance, the interfaces of an antireflection PC with the host PC and the air structure have been selected at the center of the thickness of a material film. This allowed us to use the general method of impedance matching. Consequently, the reflection at the PC–air interface was sufficiently suppressed. In addition, numerical results have revealed that our PC structure provides a high-throughput wide angular transmission window. The current nanofabrication technology for multilayers can be used for our structure. Therefore, our results nicely extend the range of potential applications of 1D PCs. For example, the signal to interference and noise ratio in optical sensor systems and communication systems can be enhanced by suppressing noise and unwanted signals coming from other directions. Based on our antireflection mechanism, one can optimize a desirable angular and frequency response for desired applications.
REFERENCES
1. J. D. Joannopoulos, S. G. Johnson, J. N. Winn, and R. D. Meade, Photonic Crystals: Modeling the Flow of Light, 2nd ed. (Princeton University, 2008).
2. H. A. Macleod, Thin-Film Optical Filters, 4th ed. (CRC Press, 2010).
3. B. E. Nelson, M. Gerken, D. A. B. Miller, R. Piestun, C.-C. Lin, and J. S. Harris Jr., Opt. Lett. 25, 1502 (2000). [CrossRef]
4. R. Ozaki, T. Matsui, M. Ozaki, and K. Yoshimoto, Appl. Phys. Lett. 82, 3593 (2003). [CrossRef]
5. K.-T. Lee, C. Ji, D. Banerjee, and L. J. Guo, Laser Photon. Rev. 9, 354 (2015). [CrossRef]
6. B. J. Lee, C. J. Fu, and Z. M. Zhang, Appl. Phys. Lett. 87, 071904 (2005). [CrossRef]
7. A. P. Raman, M. A. Anoma, L. Zhu, E. Rephaeli, and S. Fan, Nature 515, 540 (2014). [CrossRef]
8. L. Maigyte and K. Staliunas, Appl. Phys. Rev. 2, 011102 (2015). [CrossRef]
9. Y. Shen, C. W. Hsu, Y. X. Yeng, J. D. Joannopoulos, and M. Soljacic, Appl. Phys. Rev. 3, 011103 (2016). [CrossRef]
10. Y. Shen, D. Ye, I. Celanovic, S. G. Johnson, J. D. Joannopoulos, and M. Soljacic, Science 343, 1499 (2014). [CrossRef]
11. R. E. Hamam, I. Celanovic, and M. Soljacic, Phys. Rev. A 83, 035806 (2011). [CrossRef]
12. Y. Shen, D. Ye, L. Wang, I. Celanovic, L. Ran, J. D. Joannopoulos, and M. Soljacic, Phys. Rev. B 90, 125422 (2014). [CrossRef]
13. A. Alu, G. D. Aguanno, N. Mattiucci, and M. J. Bloemer, Phys. Rev. Lett. 106, 123902 (2011). [CrossRef]
14. C. Niclass, M. Soga, H. Matsubara, S. Kato, and M. Kagami, IEEE J. Solid-State Circuits 48, 559 (2013). [CrossRef]
15. T. Komine and M. Nakagawa, IEEE Trans. Consum. Electron. 50, 100 (2004). [CrossRef]
16. Y. Fink, J. N. Winn, S. Fan, C. Chen, J. Michel, J. D. Joannopoulos, and E. L. Thomas, Science 282, 1679 (1998). [CrossRef]
17. B. Shi, Z. M. Jiang, and X. Wang, Opt. Lett. 26, 1194 (2001). [CrossRef]
18. G. Liang, P. Han, and H. Wang, Opt. Lett. 29, 192 (2004). [CrossRef]
19. K.-Y. Xu, X. Zheng, C.-L. Li, and W.-L. She, Phys. Rev. E 71, 066604 (2005). [CrossRef]
20. H. Nemec, P. Kuzel, L. Duvillaret, A. Pashkin, M. Dressel, and M. T. Sebastian, Opt. Lett. 30, 549 (2005). [CrossRef]
21. Y. Takeda, H. Iizuka, S. Mizuno, K. Hasegawa, T. Ichikawa, H. Ito, T. Kajino, A. Ichiki, and T. Motohiro, J. Appl. Phys. 116, 014501 (2014). [CrossRef]
22. H. Iizuka and N. Engheta, Phys. Rev. B 90, 115412 (2014). [CrossRef]
23. K. M. Ho, C. T. Chan, and C. M. Soukoulis, Phys. Rev. Lett. 65, 3152 (1990). [CrossRef]
24. W. C. Chew, Waves and Fields in Inhomogeneous Media (Van Nostrand Reinhold, 1990).
25. B. E. A. Saleh and M. C. Teich, Fundamentals of Photonics, 2nd ed. (Wiley, 2007).