Open Access
Add to BibSonomy
Abstract
This work details the successful computational design, fabrication, and characterization of a cavity-based aluminum nanohole array. The designs incorporate arrays of aluminum nanoholes that are patterned on a dielectric-coated (SiO2 or ZnSe) aluminum base mirror plane. This architecture provided a means of exploring the coupling of the localized resonances, exhibited by the aluminum nanohole array, with the cavity resonance that is generated within the dielectric spacer layer, which resides between the base plane mirror and the nanohole array. Rigorous coupled wave analysis (RCWA) was first used to computationally design the structures. Next, a range of lithographic techniques, including photolithography, E-beam lithography, and nanosphere lithography, were used to fabricate the structures. Finally, infrared spectroscopy and scanning electron microscopy (SEM) were used to characterize the spectral and structural properties of the multilayered devices, respectively. The overall goal of this study was to demonstrate our ability to design and fabricate aluminum-based structures with tunable resonances throughout the infrared region, i.e. from the short-wave through longwave infrared regions of the electromagnetic spectrum (1.5 −12 µm).
© 2017 Optical Society of America
1. Introduction
The optimization and design of metallic nanostructures arrays continue to highlight progress being made in the nanophotonic field of research. This progress is largely due to the capability of these structures to control light-matter interactions at the nanoscale. The optical properties of these engineered materials are highly dependent on the size, shape, and periodicity of metallic and dielectric components. Hence, a plethora of designs have been explored as a means of generating and tuning various optical properties, including transmission, reflection, absorption, and emission [1–4].
One particularly intriguing design that complements the metallic nanostructure arrays is the integration of the metal arrays within a metal-dielectric-metal (MDM) stack. These multilayer thin-films typically consist of a solid metallic base plane mirror, an intermediate dielectric cavity layer with variable thickness, and a top layer consisting of the metallic grating with subwavelength features. The addition of the dielectric and metallic base plane mirror in the MDM structure creates a Fabry-Perot (FP)-like cavity that can be used to enhance and tune a strong resonant absorption [5-6]. As incident radiation impinges upon the top metallic grating layer, a resonance or absorption is generated somewhere in the UV through infrared regions, depending upon the size, shape, and periodicity of the metallic nanostructures within an array. The portion of incident radiation that is not attenuated by the array, traverses into the transparent dielectric layer and propagates towards the base plane mirror. This radiation is then reflected back through the dielectric cavity and towards the top grating surface, where it is either absorbed, transmitted back towards its original location, or reflected back to the mirror base plane where the process starts over again. At particular wavelengths the light will strongly resonate within the cavity resulting in nearly 100% absorption within the top metal grating.
The nanohole array is one of many metal-optic designs that have been explored as means of generating so called “hot spots”, or enhanced electromagnetic near fields. These hot spots may be used as a means of enhancing detection in techniques such as surface enhanced Raman spectroscopy (SERS), surface enhanced fluorescence (SEF), and surface enhanced infrared absorption (SEIRA) [7–9]. Metallic nanohole arrays are particularly attractive for these applications since they can support both localized and propagating surface plasmon resonances in the visible and UV spectra. At longer infrared wavelengths the absorption response is dominated by the FP effects. Depending upon the periodicity and size of the nanoholes, and the composition of the metal, the spectral properties of these structures may be tuned from the UV through infrared spectral regions.
To date, the vast majority of nanohole array designs have incorporated noble metals such as gold and silver. Gold and silver intrinsically exhibit low losses in the visible through near-infrared ranges, and therefore represent ideal candidates for optimizing resonances in this wavelength regime. Gold is ideal for exposure to biological media, due to its chemical stability. Conversely, silver presents toxicity issues in biological media and is therefore more suitable for environmental studies. The high cost of both gold and silver limits their use and large-scale commercialization of these sensors.
The use of aluminum represents a viable alternative to the gold and silver plasmonic nanostructures that are commonly used in plasmonic arrays [9]. First, aluminum supports resonances that occur in the UV and visible regions [9]. Second, aluminum is low-cost, abundant, and is easily processed, due to its use in a wide range of manufacturing processes. A significant consideration in using aluminum is its tendency to form an oxidative layer on its surface. Upon exposure to air, a thin layer of aluminum oxide is commonly formed. This layer provides an insulating, water-insoluble, and waterproof film that protects the inner aluminum layer. However, depending on its thickness, this thin aluminum oxide film can produce a change in the expected resonance wavelength [9].
There have been recent studies in which aluminum nanoarrays were employed for a range of applications. Wei et. al. recently employed a capped aluminum nano-slit array for use as a biosensor with tailorable Fano resonances [10]. Zhou et al employed an aluminum nano-pyramid array with tunable resonances in the UV through IR regions for the detection of carbohydrate antigen 199 [11]. Du et. al. detailed the use of an aluminum nanohole array as a transparent electrode, demonstrating a broadband transmittance of >60% [12]. A major structural feature that was omitted in all of these studies was the use of a cavity and base plane mirror to complement the tunability of the top aluminum array. Hence, in the study described herein, our specific goal was to explore the impact of a Fabry-Perot-like cavity on the resonances generated by an aluminum nanohole array.
2. Computational modeling and design
2.1 Electromagnetic modelling of aluminum nanohole arrays
The basic geometry of the cavity-based nanohole array, illustrated in Fig. 1, is composed of three layers. The bottom layer is a planar aluminum mirror of sufficient thickness (>100 nm) to ensure zero transmittance. The middle layer is composed of a low loss dielectric material with thickness, dsubstrate, and refractive index, nsubstrate. The top layer is the planar aluminum surface with a hexagonal array of circular holes removed. The distance between holes was assumed to be subwavelength and is denoted by Λ in the figure.
Fig. 1 (Top layer) Al nanohole array; (Middle layer) Transparent dielectric layer (SiO2 of ZnSe); (Bottom layer) Al base plane mirror. This “sandwiching” effect creates the necessary resonant cavity for increased use response.
Rigorous electromagnetic modeling was used to predict and design the optical properties of the cavity-based nanohole arrays. Several rigorous electromagnetic models can be used for this calculation, however, we chose the rigorous coupled wave (RCW) algorithm originally presented by Moharam and Gaylord [13]. Our specific implementation is based on the enhanced transmittance matrix approach introduced by Moharam et al. [13] and later refined by Lalanne [14] and Noponen and Turunen [15]. For the sake of brevity, we refer the reader to the references above for details on the RCW method. While accurate, the RCW method does assume the grating structure, shown in Fig. 1, is infinite in the transverse directions.
Using RCW we calculated the reflective and absorptive optical properties of the cavity-based nano-hole array. Specific parameters that were varied in the calculations included cavity thickness and dielectric material composition (ZnSe or SiO2), aluminum nanohole array dimensions (e.g. film thickness, nanohole diameter, hole-to-hole distance).
2.2 Absorption theory
To illustrate the light absorption properties of the cavity backed nanohole array, shown in Fig. 1, we conducted rigorous simulations using the RCW method. Example results from a typical simulation are shown in Fig. 2. For this example the distance between the nano-holes, Λ, was fixed at 3 μm, the hole diameter, dhole, fixed at 2 μm and the index of refraction of the substrate, nsubstrate, was assumed to be 2.41 (i.e. index of zinc selenide). The thickness of the dielectric cavity, dsubstrate, was varied from 1 μm to 7 μm. The RCW method was used to calculate the total absorption at normal incidence as a function of incident wavelength and cavity thickness. Superimposed upon those results are lines that represent the first three FP modes. Specifically, each FP mode can be described by a resonant wavelength, , given by
where is the refractive index of the incident region, is the refractive index of the dielectric cavity, is the cavity thickness, is the incident angle (e.g. is normal incidence), and is a non-zero integer that describes the mode shape. The grating and ground plane thicknesses are assumed negligible.Fig. 2 Simulated absorption intensity for normally incident plane wave illuminating a cavity backed nanohole array structure shown in Fig. 1 as a function of the wavelength and cavity thickness. Superimposed on the fig are the first three FP modes given by Eq. (1). The dashed red line was added to separate two distinct regions in which the absorption phenomenon changes from being dominated by lossy surface and guided modes (upper region) to being dominated by the FP modes (lower region).
The results in Fig. 2 indicate two interesting regions, separated by the red dashed line, in which the characteristics of the absorption phenomenon change. For wavelengths above some critical value the absorption is dominated by the FP effect. In this region the resonant wavelengths predicted by RCW are slightly red-shifted from the wavelengths predicted by (1) but follow the general trend of the FP modes. The red-shift can be accounted for by the inductive response of the metallic grating. For shorter wavelengths, however, the resonant phenomenon is more complicated due to the presence of guided and surface modes, such as lossy guided mode resonances (GMR). While the effect of the coupling between the cavity and the metallic grading is still strong in this region it is much more difficult to predict the specific response without the use of full wave simulations.
2.3 Iterative design of aluminum nanohole arrays
The resonant absorption properties of the structure shown in Fig. 1 have a complicated dependence on a number of geometrical parameters. As a result, it is unlikely that any simple analytical design equation could be derived and used to determine an optimal structure for a given desired response. Consequently, we implemented a numerical iterative design algorithm. Here the RCW method is used to calculate the full wave solution for the reflectance as a function of wavelength and polarization for a given substrate thickness, grating period and nano-hole diameter. An optimization algorithm is then used to refine the geometry until an objective function is minimized. The objective function may vary depending on the application, but in most cases we chose to minimize the total reflectance (i.e. maximize absorption) over some desired wavelength band. A number of iterative optimization algorithms could be employed including traditional derivative-based algorithms, genetic algorithms or direct pattern search algorithms. An advantage of both genetic and pattern search algorithms is that they do not require derivatives, and as a consequence work well on non-differentiable, stochastic, and discontinuous objective functions. Both simple genetic algorithms and direct pattern search algorithms were implemented and tested for the application of interest here. While both methods produced comparable results, the pattern search algorithm was often computationally less expensive.
3. Experimental methods
The generation of resonances within a given wavelength band is dependent on the feature sizes of the array, the composition of the dielectric cavity, and the thickness of the cavity. For applications in the visible and near infrared bands, the nano-holes diameters and grating periods can become quite small (i.e. << 1 µm). In the long wave infrared band (LWIR), the features sizes can be as large as several micrometers. Consequently, no single fabrication methodology is optimal across all wavelength bands. As a result, we explored three different fabrication processes (i.e. optical lithography, E-beam lithography and nanosphere lithography). The specific fabrication methods and materials used are described in the following sections.
3.1 Al nanohole structure fabrication – photolithography
Photolithography was first employed to fabricate structures that exhibited resonances in the 5-12µm region. A depiction of this process is provided in Fig. 3. Al was first deposited by electron-beam (e-beam) evaporation on the substrate in order make a back plane mirror, preventing radiation from transmitting through the Si substrate. Next, a ZnSe cavity was created atop this mirror via thermal evaporation Fig. 3(a). ZnSe was chosen as the dielectric cavity material due to its inherent transparency in this spectral region of interest. This was followed by multiple steps which involved the generation of the aluminum nanohole array. In this multi-step process, a layer of NR9-1500PY was first spin-coated onto the surface of the ZnSe Fig. 3(b), followed by a short exposure to UV light through a patterned square array mask Fig. 3(c). The mask had several square array patterns, with holes ranging from 1 to 3 µm and periods of 2-5 µm. The chosen hole diameter and period dictated the eventual dimensions of the nanohole array. After exposure to UV, the photoresist was then developed in RD-6. Next, after an appropriate solvent (acetone) was used to remove the underdeveloped resist, an array of discs remained Fig. 3(d). Next, a layer of Al was deposited over the entire disc array by means of Electron-beam (e-beam) evaporation Fig. 3(e). Finally, upon exposure to acetone and ultrasonic bath, the NR9/aluminum discs were removed, leaving behind the desired aluminum nanohole array Fig. 3(f).
Fig. 3 From left to right, starting at the top. 100nm of Al and ZnSe cavity (a), photoresist layer (b), photoresist layer after UV contact lithography exposure (c), photoresist after development (d), 100 nm of Al electron-beam (e-beam) evaporated onto SiO2 and photoresist (e), photoresist and aluminum discs removed after lift-off (f).
3.2 Al nanohole structure fabrication – E-beam lithography
In order to generate a nanohole array with smaller features, and shift resonances towards the SWIR and MWIR regions (1.5 – 5 µm), E-beam lithography fabrication was employed. A 100 nm layer of aluminum was first evaporated onto a silicon wafer in order to create a base plane mirror, which prevented the transmission of incident radiation through the device. The SiO2 cavity (200-600 nm) was then deposited onto the base plane mirror via plasma-enhanced chemical vapor deposition (PECVD) using SiH4 and N2O gases. The nanohole array was generated by electron beam lithography using a 4% dilution of polymethyl methacrylate in anisole (PMMA A4), which was developed in methyl isobutyl ketone/isopropanol (MIBK/IPA) mixture (1:3 v/v) after exposure. Once developed, 50 nm of aluminum was sputtered and lifted off to form the array on the surface of the device.
The choice of sputtering (versus electron beam evaporation) was justified when considering the negative effects of E-beam on the PMMA photoresist development. E-beam evaporation can heat the PMMA and potentially deform the PMMA during development. Conversely, sputtering did not introduce this risk and therefore provided a more suitable means of depositing a layer of aluminum.
3.3 Al nanohole structure fabrication – nanosphere lithography
Nanosphere lithography has been shown by a number of previous investigators to be a cost effective means of fabricating submicron hole arrays as well as other feature geometries [16–20]. A description of the nanosphere lithographic approach is provided schematically in Fig. 4.
Fig. 4 Nanosphere lithography fabrication process for generating a cavity-based nanohole array structure. 1) A base-plane metallic mirror is first deposited on a given substrate using physical vapor deposition (PVD). 2) A dielectric cavity layer, in this case SiO2, with a given thickness is subsequently deposited using chemical vapor deposition (CVD). 3) Next, polystyrene latex spheres are spin-coated on the SiO2 layer, forming a tightly-packed layer of nanospheres. 4) The nanospheres are then shrunk to a given size, using a reactive ion-etch (RIE) approach, e.g. oxygen plasma. 5) A thin layer of aluminum is deposited onto the spheres and into the interstitial spaces between the spheres, using electron beam evaporation. 6) Finally, the polystyrene latex spheres are dissolved in a suitable solvent such as toluene with the aid of sonication, leaving the patterned nano-hole array.
Briefly, a new 4” silicon wafer was first cleaved into 35mm square samples. The samples were cleaned in an O2 plasma for 10 minutes at high power. Silicon naturally forms a layer of SiO2 at room temperature. Utilizing this phenomena and the hydrophilic nature of silica, the polystyrene spheres were simply spin-coated onto the substrate. 490nm polystyrene spheres were drop coated onto the sample and then spun at 3000 rpm for 45 seconds. The sample was then subjected to reactive ion etch (oxygen plasma) for 2 minutes, in order to shrink the spheres and create interstitial spaces between the spheres. Next, the sample was loaded into an electron beam evaporation system. During the subsequent evaporation steps the sample holder was not rotated. A pressure of 8x10−7 torr was achieved before any deposition was initiated. A 10 nm layer of titanium was first deposited at a rate of 0.5Å/s. This layer promoted the eventual adhesion of aluminum to the substrate. Next, a 100 nm layer of aluminum was deposited at a rate of 0.5Å/s. The sample was subsequently removed from the evaporator. In order to remove the aluminum coated spheres, the sample was submerged in a beaker of toluene and placed in an ultrasonic bath for 5 minutes.
3.4 Optical characterization
The spectral features of the nanohole arrays were determined using a Fourier transform infrared (FTIR) spectrometer (Digilab Excalibur HE Series with UMA 600 Microscope). To obtain the reflectance, our FTIR spectrometer projected LWIR light through a ZnSe polarizer, focused it onto the sample with a ZnSe lens, and measured the reflected intensity with a mercury cadmium telluride (MCT) detector. The impinging beam (i.e. the beam incident on the sample following the ZnSe lens) had a spot size of approximately 0.5 mm and a half beam width of approximately 3°. This variation in incident angle was found through RCW simulations to have an insignificant effect on resonant wavelength. As a result we assumed a normally incident beam in all expected results. A Phillips LX30 environmental scanning electron microscope (ESEM) was used to image the nanohole array devices. SEM and FTIR results are provided in the subsequent section.
4. Results and discussion
4.1 SEM images of nanohole arrays
SEM images of aluminum nanohole arrays fabricated using the previously described lithographic techniques are provided in Fig. 5. In Fig. 5(a), an image of an array fabricated using photolithography is provided. Relatively large holes (average diameter = 2.28 µm) and periods (2.95 µm) were observed for this particular structure. In general, structures with these approximate nanohole array dimensions were explored as a means of generating resonances in the 5-12 µm.
Fig. 5 (a) SEM image aluminum nanohole array fabricated using photolithography. Period = 2.95 µm, average hole diameter = 2.28 µm. (b) SEM image aluminum nanohole array fabricated using E-beam lithography. Period = 0.4 µm, average hole diameter = 0.281 µm. (c) SEM image aluminum nanohole array fabricated using nanosphere lithography. Period = 0.5 µm, average hole diameter = 0.4 µm.
In Fig. 5(b), an SEM image of a typical array fabricated using E-beam lithography is provided. In general, much smaller hole diameters and periods were achieved using E-beam lithography. The average period and hole diameter for this particular structure was 0.4 µm and 0.5 µm, respectively. The goal of using E-beam lithography was to generate smaller array features in order to shift the resonances towards the SWIR and MWIR spectral regions. Figure 5(c) demonstrates a typical nanohole array fabricated using the nanosphere lithography technique. Nanosphere lithography is a much simpler and cheaper alternative to both photolithography and E-beam lithography. Hence, our goal with this technique was to demonstrate its viability in fabricating aluminum nanoholes arrays with resonances in the SWIR and MWIR spectral regions. An average period of 0.5 µm and a hole diameter of 0.4 µm was observed for this particular device.
4.2 Infrared spectral properties of cavity-based nanohole array
4.2.1. Devices fabricated with photolithography
The simulated and experimental spectral properties for the cavity-based nanohole arrays are provided in Figs. 6 and 7. In this set of experiments, ZnSe was chosen as the cavity material, due to its transparent spectral properties in the MWIR to LWIR spectral regions (5-12 µm). For the data generated in Fig. 6, an aluminum nanohole array thickness of 50 nm, a dielectric cavity thickness of 1.4 µm, and an aluminum base-plane mirror thickness of 100 nm were held constant while the impact of hole diameter, periodicity, and the size parameter (2a/Λ) were varied (here a represents the nanohole radius, and Λ represents the periodicity).
Fig. 6 Simulated and experimental spectra of cavity-based nanohole arrays fabricated using photolithography. The aluminum nanohole array thickness (50 nm), the dielectric cavity thickness (1.4 µm), and the aluminum base-plane mirror thickness (100 nm) were held constant. (a) Hole diameter = 3 µm, period = 5 µm. C. (b) Hole diameter = 2 µm, period = 4 µm. (c) Hole diameter = 2 µm, period = 3 µm. (d) Hole diameter = 1 µm, period = 2 µm.
Fig. 7 Simulated and experimental spectra of cavity-based nanohole arrays fabricated using photolithography. The aluminum nanohole array thickness (50 nm), the hole diameter (2 µm), the period (3 µm), and the aluminum base-plane mirror thickness (100 nm) were held constant. The ZnSe cavity thickness was varied as follows: (a) cavity thickness = 1.1 µm (b) cavity thickness = 1.4 µm. (c) cavity thickness = 1.7 µm. (d) cavity thickness = 2.0.
Inspection of Fig. 6 reveals that there are some clearly observable trends in both the simulated and experimental data. First, as both the hole diameter and period are decreased, the major resonances exhibited by the device are consistently blue-shifted. This observation is consistent with metal optics theory which stipulates that as the given structural features become smaller, the wavelength at which the resonance occurs becomes blue-shifted. Another observable feature in Fig. 6 is revealed when comparing Figs. 6(c) and 6(d), where an increase in size parameter produces a blue shift in the observed resonance. This result is also consistent with metal optics theory in terms of diminishing the distance between adjacent plasmonic structures. Of particular note is the magnitude of the resonances that were observed in the four test cases provided in Fig. 6. The smallest nanohole diameter of 1 µm generated the weakest resonance. Whereas the relatively larger hole diameters of 2-3 µm generated much stronger resonances, in terms of computational and experimental results. When comparing Figs. 6(a), 6(b), and 6(c), the strongest resonances were observed for those scenarios in which the size parameter was in the range 0.6 – 0.7 (i.e. for Figs. 6(a) and 6(c)). A size parameter of 0.5 yielded slightly lower magnitudes in the resonances, as observed in Fig. 6(d).
The impact of cavity thickness on spectral properties was the primary parameter that was explored for the data generated in Fig. 7. In this set of experiments, the aluminum nanohole array thickness (50 nm), nanohole diameter (2 µm), periodicity (3 µm), and base-plane mirror thickness (100 nm) were held constant. Inspection of data produced in Fig. 7 reveals that a distinct trend is readily observable. Namely, as the cavity thickness is increased from 1.1 µm to 2.0 µm, the spectral features of the resonances are steadily red-shifted. This trend is clearly observable in Figs. 7(a) – 7(c) and less so in Fig. 7(d). For the results in Fig. 7(d), we anticipated a resonance near 14µm, as shown in the computational plot. However, this resonance was not experimentally verified since the FTIR analysis was cutoff at 12 µm.
In summarizing the results provided in Figs. 6 and 7, the main parameters that dictate the resonance wavelength are the cavity thickness, nanohole diameter, and periodicity, while the size parameter played a role in the magnitude of the resonance. While, we the measured results were reasonably close to the predictions there were some clear discrepancies. We believe there are several potential reasons for the differences. One is that the index of refraction used for ZnSe (i.e. 2.41) was taken from that reported in the literature for crystalline ZnSe wafers. However, the actually index for ZnSe films could be different due to the amorphous nature of the films. Secondly, the models assumed a uniform grating with circular holes. As clearly shown in the SEM figures of Fig. 5 the actual nanohole geometries slightly deviated from the pure circular model. Lastly, since the results are very dependent on the thickness of the ZnSe cavity any errors in the film thickness measurements would shift the resonance response.
4.2.2. Devices fabricated with E-beam lithography and nanosphere lithography
Photolithography was the primary fabrication technique used to explore the impact of structural features in our study of these aluminum nanohole arrays. As a follow-on to this effort, our goal was to demonstrate our ability to generate resonances in the SWIR to MWIR spectral regions by fabricating smaller structural features, i.e. smaller nanohole diameters and periods. To achieve this goal, E-beam and nanosphere lithographic techniques were developed to fabricate the aluminum nanohole arrays with smaller structural features. Figures 8(a) and 8(b) provide computational and experimental results for devices made with the E-beam lithography and nanosphere lithography, respectively.
Fig. 8 Simulated and experimental spectra of cavity-based nanohole arrays fabricated using E-beam lithography (a) and nanosphere lithography (b).
Inspection of Fig. 8 reveals that the E-beam and nanosphere lithographic techniques both provided a means of generating resonances in the SWIR spectral region, i.e. in the 1.5-3.0 µm region. As expected, in comparison to those results generated via photolithography, smaller structural features were achieved (as demonstrated by SEM images in Fig. 5), leading to resonances that were significantly blue-shifted.
5. Conclusion
Aluminum cavity-based nanohole arrays were successfully designed computationally, fabricated, and characterized. The impact of cavity thickness on tuning the resonance of an aluminum nanohole array was the primary focus of this study. Photolithography, E-beam lithography, and nanosphere lithography were explored as a means of fabricating these devices. Nanohole diameter, periodicity of the array, and cavity thickness were all found to play a significant role in tuning the spectral properties of the devices. Photolithography was successfully used to generate devices with resonances in the LWIR (5 µm to 12 µm region). E-beam and nanosphere lithography were successfully used to generate resonances at shorter wavelengths, i.e. resonances in the SWIR spectral region. To our knowledge, this study was the first documented attempt to explore the impact of cavity thickness on tuning the resonances that are generated by aluminum-based nanohole arrays. The nanosphere lithography results also demonstrate that it is a viable technique for fabricating aluminum arrays, in addition to providing a cheaper and simpler alternative to photolithography and E-beam lithography. We anticipate that this study will serve as a basis for future studies in which cavity-based nanohole arrays will be explored as a means of enhancing detection sensitivity in SERS, SEF, and SEIRA.
Funding
Department of the Army Basic Research Program and sponsored by the Edgewood Chemical Biological Center; U.S. Army Research Office (W911NF-14-2-0104).
References and links
1. N. I. Landy, S. Sajuyigbe, J. J. Mock, D. R. Smith, and W. J. Padilla, “Perfect metamaterial absorber,” Phys. Rev. Lett. 100(20), 207402 (2008). [CrossRef] [PubMed]
2. K. Aydin, V. E. Ferry, R. M. Briggs, and H. A. Atwater, “Broadband polarization-independent resonant light absorption using ultrathin plasmonic super absorbers,” Nat. Commun. 2, 517 (2011). [CrossRef] [PubMed]
3. R. A. Pala, J. White, E. Barnard, J. Liu, and M. L. Brongersma, “Design of a plasmonic thin-film solar cells with broadband absorption enhancements,” Adv. Mater. 21(34), 3504–3509 (2009). [CrossRef]
4. D. Chanda, K. Shigeta, T. Truong, E. Lui, A. Mihi, M. Schulmerich, P. V. Braun, R. Bhargava, and J. A. Rogers, “Coupling of plasmonic and optical cavity modes in quasi-three-dimensional plasmonic crystals,” Nat. Commun. 2, 479 (2011). [CrossRef] [PubMed]
5. A. Li, S. Butun, and K. Aydin, “Large-Area, Lithography-Free Super Absorbers and Color Filters at Visible Frequencies Using Ultrathin Metallic Films,” ACS Photonics 2(2), 183–188 (2015). [CrossRef]
6. P. Zheng, M. Li, R. Jurevic, S. K. Cushing, Y. Liu, and N. Wu, “A gold nanohole array based surface-enhanced Raman scattering biosensor for detection of silver(I) and mercury(II) in human saliva,” Nanoscale 7(25), 11005–11012 (2015). [CrossRef] [PubMed]
7. A. Sanchez-Gonzalez, S. Corni, and B. Mennucci, “Surface-Enhanced Fluorescence within a Metal Nanoparticle Array: The Role of Solvent and Plasmon Couplings,” J. Phys. Chem. C 115(13), 5450–5460 (2011). [CrossRef]
8. M. Abb, Y. Wang, N. Papasimakis, C. H. de Groot, and O. L. Muskens, “Surface-Enhanced Infrared Spectroscopy Using Metal Oxide Plasmonic Antenna Arrays,” Nano Lett. 14(1), 346–352 (2014). [CrossRef] [PubMed]
9. M. W. Knight, N. S. King, L. Liu, H. O. Everitt, P. Nordlander, and N. J. Halas, “Aluminum for Plasmonics,” ACS Nano 8(1), 834–840 (2014). [CrossRef] [PubMed]
10. K. L. Lee, H. Y. Hsu, M. L. You, C. C. Chang, M. Y. Pan, X. Shi, K. Ueno, H. Misawa, and P. K. Wei, “Highly Sensitive Aluminum-Based Biosensors using Tailorable Fano Resonances in Capped Nanostructures,” Sci. Rep. 7, 44104 (2017). [CrossRef] [PubMed]
11. W. Li, Y. Qiu, L. Zhang, L. Jiang, Z. Zhou, H. Chen, and J. Zhou, “Aluminum nanopyramid array with tunable ultraviolet-visible-infrared wavelength plasmon resonances for rapid detection of carbohydrate antigen 199,” Biosens. Bioelectron. 79, 500–507 (2016). [CrossRef] [PubMed]
12. Q. G. Du, W. Yue, Z. Wang, W. T. Lau, H. Ren, and E. P. Li, “High optical transmittance of aluminum ultrathin film with hexagonal nanohole arrays as transparent electrode,” Opt. Express 24(5), 4680–4688 (2016). [CrossRef]
13. M. G. Mohoram and T. K. Gaylord, “Rigorous coupled-wave analysis of planar-grating diffraction,” J. Opt. Soc. Am. A 71(7), 811–818 (1981). [CrossRef]
14. P. Lalanne, “Improved formulation of the coupled-wave method for two-dimensional gratings,” J. Opt. Soc. Am. A 14(7), 1592–1598 (1997). [CrossRef]
15. E. Noponen and J. Turunen, “Eigenmode method for electromagnetic synthesis of diffractive elements with three-dimensional profiles,” J. Opt. Soc. Am. A 11(9), 2494–2502 (1994). [CrossRef]
16. N. Liu, M. Mesch, T. Weiss, M. Hentschel, and H. Giessen, “Infrared perfect absorber and its application as plasmonic sensor,” Nano Lett. 10(7), 2342–2348 (2010). [CrossRef] [PubMed]
17. J. Zhao, B. Frank, S. Burger, and H. Giessen, “Large-area high-quality plasmonic oligomers fabricated by angle-controlled colloidal nanolithography,” ACS Nano 5(11), 9009–9016 (2011). [CrossRef] [PubMed]
18. M. C. Gwinner, E. Koroknay, L. Fu, P. Patoka, W. Kandulski, M. Giersig, and H. Giessen, “Periodic Large-area Metallic Split-Ring Resonator Metamaterial Fabrication Based on Shadow Nanosphere Lithography,” Small 5(3), 400–406 (2009). [CrossRef] [PubMed]
19. A. Bonakdar, M. Rezaei, R. L. Brown, V. Fathipour, E. Dexheimer, S. J. Jang, and H. Mohseni, “Deep-UV microsphere projection lithography,” Opt. Lett. 40(11), 2537–2540 (2015). [CrossRef] [PubMed]
20. C. Qu and E. C. Kinzel, “Polycrystalline metasurface perfect absorbers fabricated using microsphere photolithography,” Opt. Lett. 41(15), 3399–3402 (2016). [CrossRef] [PubMed]
