Metal-insulator-metal plasmonic absorbers - OSA Publishing

Metal-insulator-metal plasmonic absorbers: influence of lattice Yiting Chen,1 Jin Dai,1 Min Yan,1 and Min Qiu1,2,∗ 1 Optics

and Photonics, School of Information and Communication Technology Royal Institute of Technology (KTH), Kista 16440, Sweden 2 State Key Laboratory of Modern Optical Instrumentation and Institute of Advanced Nanophotonics, Department of Optical Engineering, Zhejiang University, Hangzhou 310027, China ∗ [email protected]

Abstract: We experimentally demonstrate three kinds of metal-insulatormetal based plasmonic absorbers consisting of arrays of gold nanodisks distributed in different lattices, including square, triangular and honeycomb lattices. It’s found that resonances originated from localized surface plasmon undergo little changes with respect to different lattice distributions of the nanodisks. The interparticle coupling results in a minor bandwidth broadening of the fundamental mode. Different from square- and triangular-lattice absorbers, honeycomb-lattice absorber possesses a unique red-shifting (with respect to incident angles) narrow-band high-order mode, which originates from coupling of incident light to propagating surface plasmon polariton (SPP) waves. Similar high-order mode can also be generated in square-lattice absorber by increasing the period so that a smaller reciprocal lattice vector can be introduced to excite the SPP mode. Furthermore, we show that two types of resonances can interact and create Fano-type resonances. The simulation results agree well with the experimental results. © 2014 Optical Society of America OCIS codes: (240.6680) Surface plasmons; (250.5403) Plasmonics; (220.3740) Lithography.

References and links 1. T. Maier and H. Brckl, “Wavelength-tunable microbolometers with metamaterial absorbers,” Opt. Lett. 34(19), 3012–3014 (2009). 2. Y. Wang, T. Sun, T. Paudel, Y. Zhang, Z. Ren, and K. Kempa, “Metamaterial-plasmonic absorber structure for high efficiency amorphous silicon solar cells,” Nano Lett. 12(1), 440–445 (2012). 3. N. Landy, C. Bingham, T. Tyler, N. Jokerst, D. Smith, and W. Padilla, “Design, theory, and measurement of a polarization-insensitive absorber for terahertz imaging,” Phys. Rev. B 79(12), 125104 (2009). 4. X. Liu, T. Tyler, T. Starr, A. Starr, N. Jokerst, and W. Padilla, “Taming the blackbody with infrared metamaterials as selective thermal emitters,” Phys. Rev. Lett. 107(4), 045901 (2011). 5. J. Mason, S. Smith, and D. Wasserman, “Strong absorption and selective thermal emission from a midinfrared metamaterial,” Appl. Phys. Lett. 98, 241105 (2011). 6. N. Landy, S. Sajuyigbe, J. Mock, D. Smith, and W. Padilla, “Perfect metamaterial absorber,” Phys. Rev. Lett. 100, 207402 (2008). 7. H. Tao , C. Bingham, A. Strikwerda, D. Pilon, D. Shrekenhamer, N. Landy, K. Fan, X. Zhang, W. Padilla, and R. Averitt, “Highly flexible wide angle of incidence terahertz metamaterial absorber: design, fabrication, and characterization,” Phys. Rev. B 78(24), 241103 (2008). 8. A. Tittl, P. Mai, R. Taubert, D. Dregely, N. Liu, and H. Giessen, “Palladium-based plasmonic perfect absorber in the visible wavelength range and its application to hydrogen sensing,” Nano Lett. 11(10), 4366–4369 (2011). 9. Z. Jiang, S. Yun, F. Toor, D. Werner, and T. Mayer, “Conformal dual-band near-perfectly absorbing mid-infrared metamaterial coating,” ACS Nano 5(6), 4641–4647 (2011).

#222048 - $15.00 USD (C) 2014 OSA

Received 3 Sep 2014; revised 10 Nov 2014; accepted 13 Nov 2014; published 4 Dec 2014 15 Dec 2014 | Vol. 22, No. 25 | DOI:10.1364/OE.22.030807 | OPTICS EXPRESS 30807

10. J. Wang, Y. Chen, J. Hao, M. Yan, and M. Qiu, “Shape-dependent absorption characteristics of three-layered metamaterial absorbers at near-infrared,” J. Appl. Phys. 109, 074510 (2011). 11. M. Yan, J. Dai, and M. Qiu, “Lithography-free broadband visible light absorber based on a mono-layer of gold nanoparticles,” J. Opt. 16, 025002 (2014). 12. M. Pu, Q. Feng, C. Hu, X. Luo, “Perfect absorption of light by coherently induced plasmon hybridization in ultrathin metamaterial film,” Plasmonics 7, 733–738 (2012). 13. 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). 14. B. Luk’yanchuk, N. Zheludev, S. Maier, N. Halas, P. Nordlander, H. Giessen, and C. Chong, “The Fano resonance in plasmonic nanostructures and metamaterials,” Nat. Mater. 9, 707–715 (2010). 15. Y. Chen, J. Dai, M. Yan, and M. Qiu, “Honeycomb-lattice plasmonic absorbers at optical frequency: anomalous high-order resonance,” Opt. Express 21(18), 20873–20879 (2013). 16. E. Palik, Handbook of Optical Constants of Solids (Academic, 1985). 17. P. Johnson, and R. Christy, “Optical constants of the noble metals,” Phys. Rev. B 6(12), 4370–4379 (1972). 18. J. Storhoff, A. Lazarides, R. Mucic, C. Mirkin, R. Letsinger, and G. Schatz, “What controls the optical properties of DNA-linked gold nanoparticle assemblies?,” J. Am. Chem. Soc. 122(19), 4640–4650 (2000). 19. T. Jensen, L. Kelly, A. Lazarides, and G. Schatz, “Electrodynamics of noble metal nanoparticles and nanoparticle clusters,” J. Clust. Sci. 10(2), 295–317 (1999). 20. T. Kelf, Y. Sugawara, R. Cole, and J. Baumberg, “Localized and delocalized plasmons in metallic nanovoids,” Phys. Rev. B 74(24), 245415 (2005).

1.

Introduction

In recent years, great efforts have been devoted to the research of plasmonic metamaterial absorbers in the interests of their versatile applications, such as microbolometers [1], solar cell [2], imaging [3], thermal emitter [4, 5] and so on. Since the first experimental demonstration of metamaterial absorber based on electric ring resonator and cut wire media structure working on microwave wavelength range [6], a lot of metal-insulator-metal (MIM) based plasmonic absorbers have been realized, covering different wavelength regimes [7–11]. Valuable research has also been done on how to design and realize absorbers with wide-angle perfect absorption [12, 13]. However, most MIM based plasmonic absorbers consist of simple lattice structures in the top metal layer, such as 1D grating, or square lattice, because these structures are designed to generate strong absorption due to their fundamental mode with the nature of localized surface plasmon, owning to coupling between single particle and substrate reflector [6–9]. Less attention has been paid to high-order mode, which can be related to the particle lattices, or even the interaction between the fundamental and high-order mode. To our concern, study on the influence of different lattices on the absorption characteristics of such absorber can lead to deeper understanding of the interaction between light and metallic nanostructures. Such knowledge is helpful to achieve better control of realizing diversified absorption characteristics of the absorbers. Hereby, we present MIM based metamaterial absorbers with arrays of gold nanodisks on top layer distributed in different lattices, including square, triangular and honeycomb lattices. Comparison between absorbers with same particles but different lattices is carried out, and common points and differences will be investigated and discussed about. Especially, a narrow-band redshifting high-order mode is observed. Afterwards, we study absorbers with different particle sizes or different lattice constants. Furthermore, it is found that Fano resonance can be realized and controlled in square-lattice absorbers with different lattice constants. Due to its characteristic features of steep dispersion and unconventional asymmetric profile, Fano resonance has been an interesting topic and promises applications in sensors, switching, lasing and so on [14]. Finally, we sum up the results and give the conclusions.

#222048 - $15.00 USD (C) 2014 OSA


x

(a)

(b)

y

-z -y x 180 nm

30 nm 28 nm 80 nm 200 nm

Fig. 1. (a) Geometric diagram of one unit of the metamaterial absorber with gold nanodisks on top layer. (b) SEM images of the metamaterial absorber with square-, triangular- and honeycomb-lattice gold nanodisks. For all the three kinds of absorbers, they consist of the same sized nanodisks (with a diameter (φ ) of 180 nm) and the distances between adjacent nanodisks are all the same as 310 nm.

2.

Experimental and simulation results

Figure 1(a) presents the unit structure of the three kinds of absorbers: the bottom two films are 80 nm thick gold and 28 nm thick alumina respectively, and the top layer is an array of gold nanodisks with the thickness of 30 nm and diameter of 180 nm. Beneath both layers of gold, one layer of 4-nm-thick titanium is deposited to enhance the adhesion between the gold and dielectric layer. Figure 1(b) shows the SEM images of the absorbers with square, triangular and honeycomb lattices respectively, which all have the same distance of 310 nm between two nearest-neighbour nanodisks. To keep the uniformity of the gold particles, all the absorbers discussed in this paper are fabricated on the same substrate. These absorbers are fabricated by the standard process [15]. Here is a brief introduction of the fabrication process. First the gold and alumina films are deposited onto a silica substrate by e-beam evaporation, then the nanodisk arrays of different lattices are patterned onto the resist (Zep520A) by means of electron beam lithography (EBL). After development, the top layer gold is deposited onto the sample. Finally by means of liftoff to remove the residue resist and gold, the absorbers are realized. Figure 2 illustrates the measured absorption spectra of the absorbers over a wide incident angle range for both polarizations in the yz plane. First, for the TE mode (Figs. 1(a)–1(c) as described in [15]), all three kinds of absorbers share strong angle-insensitive absorption at around 1140 nm, and the average bandwidths (or FWHM in short for Full Width at Half Maximum) of this absorption peak for square-, triangle-, honeycomb-lattice absorbers are 220, 260 and 180 nm respectively. Here, we calculate the bandwidths of the experimental results by averaging the bandwidths of the fundamental mode from 0 to 60◦ , so that the measurement error can be minimized. Besides, there is no very strong high-order mode in the spectra for the TE mode, which will be further investigated in the simulation part. As for the TM mode (Figs. 1(d)–1(f) as described in [15]), the absorbers also exhibit almost perfect light absorption ability at around 1140 nm attributed to the fundamental resonance, with the average bandwidths of 230, 250 and 180 nm for the square, triangular and honeycomb lattice absorbers respectively, and the main difference appears in the high-order modes between honeycomb lattice absorber and the other two kinds of absorbers. Only the honeycomb-lattice absorber possesses a strong angle sensitive, narrow-band high-order mode, which exhibits noticeable red-shift with increasing incident

#222048 - $15.00 USD (C) 2014 OSA


Fig. 2. Measured absorption spectra of the absorbers in different lattices and polarizations at the yz incident plane: TE mode for square (a), triangular (b) and honeycomb (c) lattice absorber. TM mode for square (d), triangular (e) and honeycomb (f) lattice absorber.

angle. This red-shifting high-order mode can be interpreted as reverse propagating surface plasmon polariton (SPP) mode, which we have explained in our previous published paper [15]. For all three kinds of absorbers, their absorption spectra share one similar high-order peak at around 666 nm when the incident angle is larger than 40◦ . To confirm the experimental results, we carried out the simulation on these absorbers of the same parameters by means of the commercial software COMSOL MULTIPHYSICS. In the simulations, the dispersive permittivity of alumina is obtained from Paliks book [16], and that of gold from the data measured by Johnson and Christy [17]. We calculated the absorption spectra of the three kinds of absorbers for both polarizations and over a wide range of incident angles for the yz incident plane, which are shown in Fig. 3. The simulated absorption spectra agree well with the experimental results (Fig. 2). Firstly, the fundamental mode maintains strong absorption at around 1140 nm over a broad incident angle range for both polarizations and all three kinds of absorbers. For TE mode, the FWHMs at normal incidence for square-, triangle-, and honeycomb-lattice absorbers are 280, 285 and 228 nm respectively, while those for TM mode are 267, 285 and 227 nm respectively. On one hand, it confirms the experimental results that the bandwidth of fundamental mode for honeycomb lattice absorber is the smallest, and that for triangular lattice absorber is the largest; on the other hand, the bandwidth from the simulation results is larger than the counterpart from experimental results, which we assume that this is because the interparticle couplings between the fabricated particles is not so strong

#222048 - $15.00 USD (C) 2014 OSA


(a) Square lattice, E⊥Syz

(d) Square lattice, H⊥Syz

(b) Triangular lattice, E⊥Syz

(e) Triangular lattice, H⊥Syz

(c) Honeycomb lattice, E⊥Syz

(f ) Honeycomb lattice, H⊥Syz

Fig. 3. Simulated absorption spectra of the absorbers with the particle size of 180 nm in different lattices and polarizations at the yz incident plane: TE mode for square (a), triangular (b) and honeycomb (c) lattice absorber. TM mode for square (d), triangular (e) and honeycomb (f) lattice absorber.

as the simulated situation due to the imperfect shape of the fabricated particles. While for the high-order modes for TM polarization, only honeycomb lattice absorber possesses a redshifting narrow-band high-order mode, and all three kinds of absorbers has a high-order mode at 640 nm at large incident angles. Generally speaking, the results about high-order modes also have good agreement with the experimental counterparts. In summary, firstly, the three kinds of absorbers mainly have two common points: one is that they all exhibit almost perfect absorption at 1140 nm from the fundamental resonance, and the other is that they all have a high-order mode at about 640 nm (at around 666 nm for the experimental results). Secondly, there are mainly two differences between the absorbers with different lattices: one is that the bandwidths of their fundamental modes are different, and the other is that only honeycomb lattice absorber has a red-shifting high-order mode with respect to incident angle. In the following section, we will discuss about these common points and differences. According to both experimental and simulation results, the difference of nanodisk distribution introduces no obvious difference to the positions of the fundamental mode (at about 1140 nm) for both polarizations and the high-order mode at about 640 nm (at around 666 nm for the experimental counterpart) at large incident angle for the TM mode. The stability of the fundamental and high-order modes can be attributed to their resonance nature, which depends on the localized surface plasmon resonance between the gold nanodisk and its image counterpart in the bottom gold film, and little on inter-particle coupling. Even though the absorbers have different densities of gold nanodisks in xy plane, they all possess almost perfect absorption at 1140 nm, and this demonstrates large absorption cross section of the gold nanodisks. As for the different bandwidths of the fundamental mode for the three kinds of absorbers, it can be attributed to inter-particle coupling between adjacent gold nanodisks. As shown in Fig. 1(b) by the yellow arrows, we see that for the triangular lattice distribution, every nanodisk has 6 nanodisks sitting closely around it with the same distance of 310 nm, while for the square-

#222048 - $15.00 USD (C) 2014 OSA


Fig. 4. Measured multi-angle absorption spectra of the honeycomb lattice absorbers for the case of H⊥Sxz with the same lattice constant and different particle sizes, with the radius of 80 nm (a) and 90 nm (b) respectively. Insets are the SEM images of the honeycomb lattice absorbers.

and honeycomb-lattice cases, the number is 4 and 3 respectively. Therefore, the nanodisk in the triangular lattice absorber has strongest inter-particle resonance, thus having largest bandwidth for the fundamental mode. On contrary, the honeycomb lattice absorber has smallest one. This phenomenon has been observed and explained in many papers [18, 19] about clusters of metallic nanoparticles that aggregating of nanoparticles leads to larger damping due to increasing volume of all particles and broadening of bandwidth. Meanwhile, the red-shifting high-order SPP mode of the honeycomb lattice absorber is aroused by the coupling of propagating wave with its wave vector added by a reciprocal lattice vector [20]. Thereby, this propagating SPP mode is mainly determined by the lattice constant or period of the particles in particular direction. To further confirm this theory, we present Fig. 4, the measured absorption spectra of two honeycomb lattice absorbers, whose only difference is their particle size, with the diameter of 160 (Fig. 4(a)) and 180 nm (Fig. 4(b)) respectively. As the particle size decreases from 180 nm to 160 nm, the main absorption peak shifts from 1140 to 1050 nm, and the absorption drops several percent but still stays above 95% at a wide incident angel range, which indicates that the particles (φ = 160 nm) interact with light close to the geometric cross section in xy plane. As shown in Fig. 4, the red-shifting high-order mode experiences no obvious movement, which manifests that this mode is mainly dependent on the lattice constant or period of the particles in particular direction. It is also possible for the square lattice absorber to exhibit such kinds of high-order mode, so long as we increase the period of the square lattice to introduce a proper reciprocal lattice vector to excite the SPP mode. Therefore we calculated the absorption spectra of square lattice absorbers with different periods for the TM mode at 60◦ incident angle, with the same gold nanodisk diameter (180 nm). The result is shown in Fig. 5(a). We can see that when the period is 310 nm, no such narrow-band peak appears. But when the period increase to 465 nm, there is a narrow-band mode at around 883 nm, which moves to 960 nm as the period changes to 510 nm, with an FWHM of about 20 nm. More interestingly, with the period going up to 570 and 620 nm, this high-order mode interferes with the fundamental mode, resulting in a sharp asymmetric dip in the fundamental absorption band at 1095 and 1180 nm respectively, which

#222048 - $15.00 USD (C) 2014 OSA


(a)

(b)

Fig. 5. Simulated (a) and experimentally measured (b) absorption spectra of the TM mode for the square lattice absorbers with different periods at 60◦ incident angle with the same particle size of φ = 180 nm.

can be interpreted as a Fano Resonance [14]. To confirm the simulation results, we fabricated the square-lattice absorbers with the same parameter used in the simulation, and the measured absorption spectra is demonstrated in Fig. 5(b). Good agreement can be found between the simulation and experimental results: the narrow-band high-order mode appear in the absorbers with the period of 465 nm and 510 nm, and the Fano Resonance appears in the absorbers with the period of 570 nm and 620 nm. Thus by increasing the period in square-lattice absorber, the narrow-band high-order mode is introduced. At the same time, it should also be noticed that absorption of the fundamental mode also undergoes noticeable drop due to decreasing density of the nanodisks. Hereby, we can see the advantage of honeycomb lattice absorber, which not only exhibits rich resonance information at different incident angle and polarizations due to its special lattice symmetry, but also possesses strong absorption in the fundamental mode. 3.

Conclusion

In summary, we demonstrate both experimental and simulation results on the MIM based plasmonic absorbers comprised of periodic arrays of gold nanodisks in different lattice distributions. It is found that the resonance modes stemmed from localized plasmon resonance exhibit little sensitivity to the lattice distribution, except that a denser package of gold particles can broaden the bandwidth of the fundamental resonance. While a red-shifting SPP mode is detected for the honeycomb lattice absorber due to its different symmetry and periodicity, and this mode is related to reciprocal lattice vector of the honeycomb lattice, rather than the particle size. However, this SPP mode can also be generated by other lattice absorber by means of increasing the period to introduce proper reciprocal lattice vector, at the cost of a drop of the absorption of the fundamental mode. Compared to square and triangular lattice absorber, honeycomb lattice absorber exhibits richer absorption characteristics, thus revealing the possibility of achieving more potential applications. Meanwhile, due to the coherent interference of the localized plasmon resonance and SPP mode between the fundamental mode and high-order mode, Fano resonance can be realized with appropriate lattice constant and particle size. Therefore, by tailoring the particle size, lattice and lattice constant, we can generally control the absorption characteristics of the absorber, not only decide the position and intensity of the fundamental and high-order modes, as well as the interaction between them. Thus, we can either introduce Fano resonance at desired wavelength, or avoid Fano resonance to achieve higher absorption efficiency. With the steep window in the Fano lineshape, the MIM based plasmonic absorber also possesses the potential to work

#222048 - $15.00 USD (C) 2014 OSA


as a reflective filter, or a reflective polarizer. Besides, as Fano resonances are based on resonances between two or more oscillators, they are usually sensitive to the changes in geometry and local environment: small perturbations can result in remarkable resonance or lineshape change [14]. This property also makes the absorber attractive to a range of applications, such as biological sensor, switch and so on. Acknowledgment This work is supported by the Swedish Foundation for Strategic Research (SSF), the Swedish Research Council (VR), and VR’s Linnaeus center in Advanced Optics and Photonics (ADOPT).

#222048 - $15.00 USD (C) 2014 OSA
