We demonstrate that multi-band coherent perfect absorption can be achieved at infrared frequencies by a metasurface in which four-sized columnar metal patches are separated by a dielectric layer in a unit cell. The absorption bandwidth is enhanced by three times compared with single-band absorption while high absorbance is maintained. The coherent perfect absorption is polarization-independent and can be independently modulated at each resonant frequency by tuning the phase difference of two coherent incident beams. Moreover, the resonant frequency is sensitive to the radius of the columnar patch, and thus a wide coherent perfect absorption frequency range can be obtained by adjusting the radius. Through optimizing the structural parameters, nearly perfect absorption at oblique incidence for both TE and TM polarizations are achieved. The optimized metasurface can be used as a beamsplitter at oblique incidence.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Light absorption plays an important role in many applications, including radiometer, stealth technology, thermal emission, optical switch and modulator [1–12]. For traditional metal conductive film, due to impedance mismatch between metal and free space, its absorption in the microwave range is very low . Although reducing the thickness of film can improve the match performance and enhance the absorption efficiency, its maximum absorption is limited to 50%, which cannot be broken under the ultra-thin approximation .
In 2008, Landy et al.  proposed the concept of metamaterials perfect absorbers based on the electromagnetic resonance, and about nearly 100% perfect absorption of the electromagnetic wave can be achieved in a specific frequency. Electromagnetic metamaterials are artificially constructed structures or composite materials with extraordinary electromagnetic properties that are not available in natural materials, and attract more and more attention in recent years [14,15]. As 2D version of metamaterials, metasurfaces of deeply subwavelength thickness can dramatically modify the amplitude, phase and polarization of light [16–21]. Many metamaterial absorbers have been widely researched from visible to infrared frequencies [1,2,5–11]. However, the metamaterial absorbers have a typical drawback in the working wavelength and absorption efficiency which are determined by the original structure and difficult to control flexibly. This drawback seriously impairs its performance as an optical switch or a modulator.
The way of the coherent perfect absorption (CPA) could provide a solution for the above mentioned problem, in which the absorption of light is controlled by two coherent beams with equal intensities propagating in opposite directions [22–32]. With the influence of absorption and interference, the absorption of light can be dynamically tuned from 0% to 100% by simply changing the relative phase of two incident beams, and thus the metasurface can achieve the transitions from transparent states to opaque states [23,24,33,34]. A typical coherent perfect absorber is a FP dielectric cavity in which all the incident energy can be captured and dissipated inside the cavity. Up to now, many structures are presented to achieve CPA, including thin films [25,28,30,34–36], grating , metal-insulator-metal structure . However, as a time-reversal laser , many coherent perfect absorbers of above metasurfaces and metamaterials only work at a single frequency, and thus the absorption bandwidth is narrow and the expanding to multi-band or broadband is difficult. The narrow bandwidth severely limits its broadband applications, such as solar energy absorption for thermal photovoltaic  and microwave absorption for electromagnetic stealth  that require a wide absorption band. Therefore we urgently need to find a way to achieve multi-band or broadband coherent perfect absorption.
Until now, many efforts have been done to achieve this goal, including thin resistive sheet with ultrabroadband absorption spectrum which was originally proposed in  and later demonstrated in . Others structures include a metasurface with four different periodically arranged meta-atoms , and multilayered split rings . But these absorbers still cannot satisfy the multiband applications. For instance, because the bandwidth of resistive sheet is always very broad, it cannot retain the light transparent of other frequencies that are not required. Moreover, the bandwidth of metasurface proposed in  is still very narrow. The structure of multilayered split rings is quite complicated which increases the difficulty of industrial manufacturing.
In this paper, we propose a metasurface composed by the unit cells in which four-sized column patches are separated by a dielectric layer that can achieve CPA in four separated frequencies. By simply adjusting the structure parameters, we can obtain a broadband CPA of which absorption rates at these frequencies are greater than 90% and can be flexibly tuned by phase modulation. The absorption performance of oblique incidence is also studied. When increasing the incidence angle, the absorption is slightly reduced for transverse electric (TE) polarization but it is not varied for transverse magnetic (TM) polarization. In order to improve its performance as a beam splitter, we optimize the structure parameters for the case of oblique incident and obtain strong absorption for both TE and TM polarizations at the incident angle of 45°. The CPA frequencies of our metasurface can be tuned within a wide frequency range and extended easily to the other desired frequencies.
As shown in Fig. 1, the metasurface suggested in this paper consists of three layers where two metallic patterned films with thicknessare separated by a dielectric spacer with thickness. Gold is used as metallic material and experimental value of dielectric constant is applied . At a frequency of 100 THz, the real part of the dielectric constant is −339.13, and the imaginary part is 60.09, respectively. The dielectric spacer is chosen as Al2O3 and the permittivity is 2.92 at frequencies around 100 THz. The period of each unit cell is. The gold film in each unit cell is etched into four-sized column patches with radiusand the thickness. Two counter-propagating coherent beams alongwith the same amplitudes and different phases are used as excitations. Our simulation and calculation are based on commercial software Computer Simulation Technology Microwave Studio (CST MWS) and the corresponding formula in , respectively.
3. Single-band CPA
For illustration, we first calculate the coherent absorption of a metasurface in which the column patches have the same radius in a unti cell (). For the metasurface in our paper, there are many parameters that can be changed, such as gold film thickness, dielectric layer thickness and the radius of column patches. For simplicity, the thickness of the gold film remains 40nm throughout the optimization. For the single-band CPA, we keep the radius of column patches constant and optimize the thickness of the dielectric layer to achieve 100% perfect absorption. The optimized parameters of and are used to satisfy the condition of CPA and the phase difference of the two coherent beams is set tofor simplicity. In the metamaterial perfect absorber theory, both reflection and transmission coefficients must be weakened () , but the CPA is completely different which requires meet two conditions: (i) forward and backward waves are of the same amplitudes and (ii) reflection and transmission coefficients have same amplitudes () andphase difference withbeing an arbitrary integer when illuminated by a single beam . The absorption spectra is shown in Fig. 2(a) and from it, we can see that the maximum absorption is 99.9% at.
The reason why the absorption peak occurs at the frequency of 109.5 THz can be understood from Figs. 2(b) and 2(c), which show the amplitudes of reflection and transmission coefficients and the corresponding phases when a single beam incident, respectively. We can see that the reflection coefficient is equal to the transmission coefficient at two frequencies of 109.5 THz and 193.5 THz where the first condition of CPA is satisfied. Only at the frequency of 109.5 THz, where the phase difference is 1.479° closed to 0, the second condition of CPA is also satisfied and a maximum absorption is obtained. When, the phase difference is 260.73° and does not meet the second CPA condition, therefore there is no obvious absorption peak even though . Changing the radius from, the simulation results are shown in Fig. 2(d). We can find that when the radius is reduced, the absorption peak moves toward high frequencies while moving to the low frequencies when the radius increases. This relationship can provide a useful reference for us to design multi-band CPA devices. In the following, we will study the size effects of the columnar patches on absorption spectra.
4. Multi-band CPA
When we add the number of columns with different radius in a unit cell, the number of absorption peaks will also increase which can be seen in Figs. 3(a)-3(c). Now we consider a metasurface composed of four-sized column patches in a unit cell with respectively. Because the absorption peak frequency is affected by the radius of the metal patches, the simulations on the combinations of different metal patches radius in a unit cell are performed repeatedly to achieve identifiable multiple peaks, and thus the optimized radius parameters were obtained. To meet the condition of CPA, the thickness of the dielectric layer is tuned to 235nm through the same optimization process as single-band CPA. Their coherent absorption spectrum is shown in Fig. 3(d). Nearly perfect coherent absorption occurs at the frequencies of , , and .
The physical mechanism for the appearance of the four absorption peaks can be explained as follows. The unit cell is composed by four subunits. At a specific frequency of each subunit, the current in the upper and lower metal patches flows in the opposite direction, which forms a current loop and produces a strong magnetic resonance coupling with the magnetic field of the incident light [38, 43]. Here, the magnetic resonance results in a significantly enhanced of the magnetic field in the dielectric layer so that light is completely absorbed by the metaurface, which can be seen from Figs. 3(e)-3(h). As described above, metal patches have different radius corresponding to different magnetic resonance response frequencies. Each magnetic resonance is independent of the others, and the total light response on the metasurface is a linear superposition of four-sized subunits.
For the previous metasurface in , its response frequencies are 23.86 THz, 25.2 THz, 26.36 THz and 27.47 THz, respectively. The response range is 3.61 THz, this narrow band limits its potential application. Our design can extend the range to 24.6 THz and maintain a high absorption at each frequency. Despite multi-layer structure in  can also achieve multi-band absorption, its manufacturing process is complex. Our structure is superimposed by horizontal, which can be simplified in the process and maintain it ultra thin.
5. Broadband CPA
In our suggestion metasurface, based on the physical mechanism of multi-band CPA, broadband CPA can be obtained when we change the radius of the four metal column patches in a unit cell. Use the same optimization process as multi-band CPA, adjust the radius of the metal patches to move several absorption peaks closer together, and then optimize the thickness of the dielectric layer to obtain the perfect absorption. The broadband absorption spectrum is shown in Fig. 4(a) and the optimized parameters are and The bandwidth is enhanced as the frequency range for the absorption rate greater than 90%. Compare with the single-band absorption in Fig. 2(a) of which the bandwidth is 2.7 THz, the bandwidth of the broadband CPA is 9 THz which is expanded to more than three times. Furthermore, compared with the former absorber in , our broadband design here also increases the bandwidth by 2.5 times and always maintains more than 90% absorption within the desired frequency range.
Then we investigate the absorption spectra at different polarization angles (), and it is clear that the absorption is almost constant due to the symmetry of the structure when the polarization angle changes . This indicates that the absorber is polarization-independent at normal incidence.
As mentioned above, coherent perfect absorber has a significant advantage that when the structure is fixed, the absorbance can be tuned by changing the phase difference of two coherent input beams. The phase modulation of coherent absorption in our suggestion structure is shown in Fig. 5(a). When the phase difference varies from 0° to 360°, the absorption of the entire broadband changes at the same time. Thus we can simultaneously tune the broadband absorption rate from 0 to 1, between completely transparent state to completely absorbed state. The phase modulation of absorption for our designed metasurface is useful as a broadband absorption modulator.
In order to better show the modulation ability of our metasurface, we show the coherent absorption as a function of phase difference at two peak frequencies 106.5 THz and 112.8 THz in Fig. 5(b). The modulation depth is defined as the ratio of the maximum absorption rate and minimum value , which represents the modulation capability of the absorber. It is 38.49 and 32.89 in our structure, respectively (see Table 1).
Actually, taking into account the possible deviations in the experimental production, the thickness of the dielectric layer cannot be exactly equal to 185nm. Therefore, we investigate the effect of dielectric layer thickness varying on broadband absorption. As shown in Fig. 6(a), it is easy to find that when the thickness varies from 180nm to 205nm, the coherent absorption is not significantly reduced. The peaks are always maintained at more than 97% which can be considered to be effective from the point of view of the absorber application. In this way, we provide a useful reference range for the thickness in the experiment.
In our suggestion CPA, the dielectric material is chosen to be Al2O3 and the dielectric constant () is 2.92 in the frequency range we calculated. Then we will study the effects of different dielectric constants on absorption. From Fig. 6(b), we can see that there are still some absorption peaks when the dielectric constant varies and the only difference is the shift of the peak frequencies. In other words, the peak frequencies are dependent on the material of dielectric layer. This relationship introduces a frequency tunable capability for the metasurface we designed, and the absorption peaks can be moved to other interested frequencies.
6. Oblique Incidence
The potential applications of angle-independent absorption are also very important, and thus we study the absorption properties of our suggestion CPA under oblique incidence for both TE and TM polarizations. As shown in Figs. 7(a) and 7(b), for TM polarization, the increased incidence angle () has no significant effect on the coherent absorption performance, and the absorbance always remains greater than 96% at the resonant frequencies for the incident angle varying from to . These trends are very similar to those in other suggestion perfect metamaterial absorbers . While for TE polarization, the absorption at large incidence angle is deteriorated. When the angles are less than, the absorbance decreases slightly and remains 96%. With the increase of the incident angle beyond , the peak absorption decreases significantly and becomes 44% at .
Figures 7(c) and 7(d) are the sketch figures which are helpful for understanding the physical mechanism of the above results. As demonstrated above, light absorption is mainly due to magnetic resonance. For TM polarization, when the incident angle increases, the magnetic field component in XY plane remains unchanged and does not affect the magnetic resonance intensity. However, this is completely different for TE polarization, the magnetic field component in XY plane is weakened with the increase of the incident angle. The transform in the intensity of the magnetic field component is similar to the cosine curve. When the incident angle increases, the magnetic field intensity in XY plane is drastically reduced, and therefore the incident magnetic field can no longer effectively support magnetic resonance .
As a beamsplitter for practical application, high absorption at the incidence angle of 45° is expected. Because in the case of normal incident, the incident beam coincides with the outgoing beam, as shown in Fig. 8(a), and thus the two beams cannot effectively be separated. However, when the beams are oblique incident, for example, the incident beams are completely separated from the outgoing beams in horizontal and vertical directions on the optical path, as shown in Fig. 8(b). And in this case the metasurface can be effectively used as a splitter. Therefore we need to optimize the absorption performance at for TE polarization.
Using the same procedure as before, the geometrical parameters are optimized with . Figures 9(a) and 9(b) show the coherent absorption spectra of the optimized metasurface for TE and TM polarizations, respectively. Absorbance is greater than 90% at the frequencies from 101.7 to 107.4 THz for TE polarization at. The bandwidth is 5.7 THz which has been improved compared with the initial metasurface for which bandwidth is 3 THz. Importantly, the absorption for TM polarization is no significantly decreased. In addition, the broadband resonance frequencies are tunable. We simply reduce the radius of metal column patches to while the other parameters are unchanged, the absorption peak shifts to high frequencies for both TE and TM polarizations in Figs. 9(c) and 9(d). These trends are very similar to the absorption in a single-sized metasurface shown in Fig. 2(d), which again confirms that the broadband CPA is a linear superposition of single-band CPA. When the radius increases, the absorption peak shifts to low frequencies. This effect of radius provides a way to tune the absorption frequency. Thus we can obtain high absorption at any desired frequency using this metasurface.
It should be mentioned that coherent absorption at can also be tuned by changing the phase difference of the two incident beams. Therefore, as a splitter or absorption modulator we can use one beam to modulate the output light intensity of the other from 0 to 100%. Finally, there is an interesting thing when we extend the simulation frequency to 300 THz to calculate coherent absorption of the metasurface. At large incidence angles, other absorption peaks occur at higher frequencies around 210 THz and 218 THz [Fig. 10(a)], which are related with the interplay of localized and delocalized plasmon resonances. The mechanism can be explained using Bragg scattering theory [38,48]. The angle-dependent absorption at high frequencies can be used to achieve the modulation of the point source. The light at large incident angles is absorbed and point source just can be incident at small angles, as shown in Fig. 10(b). The modulation may provide helpful guidance for potential optical application.
In conclusion, we have designed a multi-band three-layered metasurface composed of unit cells in which four-sized metal column patches are separated by a dielectric layer and calculated coherent absorption spectra. The different radius of the four column patches in a unit cell results in four magnetic resonances at different frequencies. The resonant frequency is influenced by the radius of metal patches, and broadband absorption can be achieved by optimized parameters. The absorption is polarization independent and the bandwidth in our design is significantly enhanced compared with the single-band absorbers. This may be used on solar cells or other broadband absorbers in the future. Furthermore, by changing the phase difference of the two coherent incident beams, absorption can be modulated from 2.58% to 99.56% at 106.5 THz, 3.03% to 99.90% at 112.8 THz, respectively. Finally, we have studied the effect of the incident angle on absorption. Due to weakened magnetic resonance, the absorption is significantly reduced for TE polarization when angle increases. The optimized structure at can be used as a beamsplitter. Because of the high absorbance at large incident angles, our metasurface can also be used to modulate point source at high frequencies.
Program 973 (2013CBA01700); Sichuan Provincial Department of Education (16ZA0047); the Large Precision Instruments Open Project Foundation of Sichuan Normal University (DJGX2017050, DJGX2017051, DJGX2017052).
This work was supported by State Key Laboratory of Optical Technologies on Nano-Fabrication and Micro-Engineering, Institute of Optics and Electronics, Chinese Academy of Sciences.
References and links
2. T. V. Teperik, F. J. Abajo, A. G. Borisov, Y. Sugawara, J. J. Baumberg, M. Abdelsalam, and P. N. Bartlett, “Omnidirectional absorption in nanostructured metal surfaces,” Nat. Photonics 2(5), 299–301 (2008). [CrossRef]
3. C. Wu, Y. Avitzour, and G. Shvets, “Ultra-thin wide-angle perfect absorber for infrared frequencies,” Proc. SPIE 7029, 70290W (2008). [CrossRef]
4. V. V. Truong and B. de Dormale, “Optical absorption in overcoats of nanoparticle arrays on a metallic substrate,” Plasmonics 6(2), 195–200 (2011). [CrossRef]
5. M. Pu, C. Hu, M. Wang, C. Huang, Z. Zhao, C. Wang, Q. Feng, and X. Luo, “Design principles for infrared wide-angle perfect absorber based on plasmonic structure,” Opt. Express 19(18), 17413–17420 (2011). [CrossRef] [PubMed]
6. M. Diem, T. Koschny, and C. M. Soukoulis, “Wide-angle perfect absorber/thermal emitter in the terahertz regime,” Phys. Rev. B 9(3), 3101 (2008).
7. C. M. Watts, X. Liu, and W. J. Padilla, “Metamaterial electromagnetic wave absorbers,” Adv. Mater. 24(23), OP98 (2012). [PubMed]
12. D. Schurig, J. J. Mock, B. J. Justice, S. A. Cummer, J. B. Pendry, A. F. Starr, and D. R. Smith, “Metamaterial electromagnetic cloak at microwave frequencies,” Science 314(5801), 977–980 (2006). [CrossRef] [PubMed]
13. H. Bosman, Y. Y. Lau, and R. M. Gilgenbach, “Microwave absorption on a thin film,” Appl. Phys. Lett. 82(9), 1353–1355 (2003). [CrossRef]
17. J. K. Gansel, M. Thiel, M. S. Rill, M. Decker, K. Bade, V. Saile, G. von Freymann, S. Linden, and M. Wegener, “Gold helix photonic metamaterial as broadband circular polarizer,” Science 325(5947), 1513–1515 (2009). [CrossRef] [PubMed]
18. N. Yu, P. Genevet, M. A. Kats, F. Aieta, J. P. Tetienne, F. Capasso, and Z. Gaburro, “Light propagation with phase discontinuities: generalized laws of reflection and refraction,” Science 334(6054), 333–337 (2011). [CrossRef] [PubMed]
19. X. Luo, “Principles of electromagnetic waves in metasurfaces,” Sci China. 58(9), 594201 (2015).
20. M. Pu, X. Li, X. Ma, Y. Wang, Z. Zhao, C. Wang, C. Hu, P. Gao, C. Huang, H. Ren, X. Li, F. Qin, J. Yang, M. Gu, M. Hong, and X. Luo, “Catenary optics for achromatic generation of perfect optical angular momentum,” Sci. Adv. 1(9), e1500396 (2015). [CrossRef] [PubMed]
21. X. Li, L. Chen, Y. Li, X. Zhang, M. Pu, Z. Zhao, X. Ma, Y. Wang, M. Hong, and X. Luo, “Multicolor 3D meta-holography by broadband plasmonic modulation,” Sci. Adv. 2(11), e1601102 (2016). [CrossRef] [PubMed]
23. J. Zhang, K. F. Macdonald, and N. I. Zheludev, “Controlling light-with-light without nonlinearity,” Light Sci. Appl. 1(7), e18 (2012). [CrossRef]
24. M. Pu, Q. Feng, M. Wang, C. Hu, C. Huang, X. Ma, Z. Zhao, C. Wang, and X. Luo, “Ultrathin broadband nearly perfect absorber with symmetrical coherent illumination,” Opt. Express 20(3), 2246–2254 (2012). [CrossRef] [PubMed]
29. M. Kang, F. Liu, T. F. Li, Q. H. Guo, J. Li, and J. Chen, “Polarization-independent coherent perfect absorption by a dipole-like metasurface,” Opt. Lett. 38(16), 3086–3088 (2013). [CrossRef] [PubMed]
30. G. Ramakrishnan, G. K. Ramanandan, A. J. Adam, M. Xu, N. Kumar, R. W. Hendrikx, and P. C. Planken, “Enhanced terahertz emission by coherent optical absorption in ultrathin semiconductor films on metals,” Opt. Express 21(14), 16784–16798 (2013). [CrossRef] [PubMed]
32. C. Yan, M. Pu, J. Luo, Y. Huang, X. Li, X. Ma, and X. Luo, “Coherent perfect absorption of electromagnetic wave in subwavelength structures,” Opt. Laser Technol. 101, 499–506 (2018). [CrossRef]
33. M. Kang, Y. Chong, H. Wang, W. Zhu, and M. Premaratne, “Critical route for coherent perfect absorption in a Fano resonance plasmonic system,” Appl. Phys. Lett. 105(13), 131103 (2014). [CrossRef]
34. J. Zhang, C. Guo, K. Liu, Z. Zhu, W. Ye, X. Yuan, and S. Qin, “Coherent perfect absorption and transparency in a nanostructured graphene film,” Opt. Express 22(10), 12524–12532 (2014). [CrossRef] [PubMed]
35. X. Fang, M. Tseng, J. Ou, K. MacDonald, D. P. Tsai, and N. Zheludev, “Ultrafast all-optical switching via coherent modulation of metamaterial absorption,” Appl. Phys. Lett. 104(14), 3 (2014). [CrossRef]
36. A. L. Fannin, J. W. Yoon, B. R. Wenner, J. W. Allen, M. S. Allen, and R. Magnusson, “Experimental evidence for coherent perfect absorption in guided-mode resonant silicon films,” IEEE Photonics J. 8(3), 6802307 (2016). [CrossRef]
37. W. Zhu, F. Xiao, M. Kang, and M. Premaratne, “Coherent perfect absorption in an all-dielectric metasurface,” Appl. Phys. Lett. 108(12), 139 (2016). [CrossRef]
38. M. Pu, Q. Feng, C. Hu, and X. Luo, “Perfect absorption of light by coherently induced plasmon hybridization in ultrathin metamaterial film,” Plasmonics 7(4), 733–738 (2012). [CrossRef]
39. A. Kohiyama, M. Shimizu, and H. Yugami, “Unidirectional radiative heat transfer with a spectrally selective planar absorber/emitter for high-efficiency solar thermophotovoltaic systems,” Appl. Phys. Express 9(11), 112302 (2016). [CrossRef]
40. S. Li, J. Luo, S. Anwar, S. Li, W. Lu, Z. H. Hang, Y. Lai, B. Hou, M. Shen, and C. Wang, “An equivalent realization of coherent perfect absorption under single beam illumination,” Sci. Rep. 4(1), 7369 (2014). [CrossRef] [PubMed]
41. S. Li, J. Luo, S. Anwar, W. Lu, Z. Hang, Y. Lai, B. Hou, M. Shen, and C. Wang, “Broadband perfect absorption of ultrathin conductive films with coherent illumination: super performance of electromagnetic absorption,” AIP Adv. 91(10), 773–775 (2014).
42. D. Xiao, K. Tao, Q. Wang, Y. Ai, and Z. Ouyang, “Metasurface for multi-wavelength coherent perfect absorption,” IEEE Photonics J. 9(1), 6800108 (2017). [CrossRef]
43. G. Nie, Q. Shi, Z. Zhu, and J. Shi, “Selective coherent perfect absorption in metamaterials,” Appl. Phys. Lett. 105(20), 201909 (2014). [CrossRef]
44. E. D. Palik, “Aluminum Oxide (AL2O3),” in Handbook of Optical Constants of Solids (Academic, 1991).
45. R. Feng, J. Qiu, L. Liu, W. Ding, and L. Chen, “Parallel LC circuit model for multi-band absorption and preliminary design of radiative cooling,” Opt. Express 22(S7Suppl 7), A1713–A1724 (2014). [CrossRef] [PubMed]
46. H. Tao, C. M. Bingham, A. C. Strikwerda, D. Pilon, D. Shrekenhamer, N. I. Landy, K. Fan, X. Zhang, W. J. Padilla, and R. D. Averitt, “Highly flexible wide angle of incidence terahertz metamaterial absorber: Design, fabrication, and characterization,” Phys. Rev. B 78(24), 241103 (2008). [CrossRef]
47. Y. Todorov, L. Tosetto, J. Teissier, A. M. Andrews, P. Klang, R. Colombelli, I. Sagnes, G. Strasser, and C. Sirtori, “Optical properties of metal-dielectric-metal microcavities in the THz frequency range,” Opt. Express 18(13), 13886–13907 (2010). [CrossRef] [PubMed]
48. T. A. Kelf, Y. Sugawara, R. M. Cole, J. J. Baumberg, M. E. Abdelsalam, S. Cintra, S. Mahajan, A. E. Russell, and P. N. Bartlett, “Localized and delocalized plasmons in metallic nanovoids,” Phys. Rev. B 74(24), 245415 (2006). [CrossRef]