Nanohole-based biosensors are some of the most promising next generation biosensor platforms thanks to their size selectivity and position dependent sensing properties. For practical applications, large-area fabrication with properly controlled dimensions and sensor performance is essential. Here, we investigate the size controllability of hexagonally ordered plasmonic nanohole arrays produced through self-assembly of mask colloid nanoparticles, and analyzed the hole-size dependent sensor properties by electromagnetic simulations. Size-reduction of the colloids to about half of their original size is the smallest achievable hole size, which roughly corresponds to the optimal diameter for the refractive index sensing. The field enhancement tuning is also discussed.
© 2016 Optical Society of America
Plasmonic materials with nanostructured surface have been attracting large attention in the last few decades both from a scientific interest in their sensitive optical responses and also for their potential for utilization, e.g. optoelectronic devices, biosensors, nano-scale waveguides, and light emission enhancement [1–4 ]. Among these possibilities, plasmonic biosensors seem most promising and realistic for practical applications considering their advantages over existing technologies [5,6 ]. The strongest advantage of the plasmonic biosensor is the use of nano-structured sensor surfaces instead of planar ones. Surface nano-structuring enables specific detection of analytes, taking advantages of e.g. local field enhancement, directed medium flow, size filtering, and local heating [7–9 ]. For label-free sensing, such multi-functional features strengthen the selectivity and expand on the possible sensing targets, since such label-free methods solely depend on the change of surface refractive index (RI). Numbers of different RI sensor platforms using plasmonic structures have been proposed; however, not all of them are durable enough for practical use, mainly because of fabrication complications and issues with reproducibility.
Nanohole-based biosensors, which can be produced through simple and easily reproducible processes, have been proven to realize such functional sensing. For instance, particle capturing, membrane sensing, virus sensing, and massive parallelization of label-free sensing have already been demonstrated by many different research groups [5,7,8,10–12 ]. Furthermore, the conductivity of the metal film, not easily achieved in particle systems, offers electric functionality, opening up for possibilities such as combination with electrochemistry [11,13 ]. The optical properties of the nanohole sensor can easily be tailored by changing parameters like the metal layer thickness and array period, which holds extra attractions for research applications. To produce nanholes, colloidal lithography is probably the simplest method, especially for short-range ordered (SRO) hole arrays [14–17 ]. Compared to top-down lithographic processes, such as focused ion beam and e-beam exposures, bottom-up methods are more favorable in terms of low-cost and large area fabrication. The most efficient hole array configuration for optical sensing is a close packed hexagonal array, as it has the highest nanohole density and well defined periodicity for the resonance [15,18 ]. For flow through sensing configuration with nanopores, large hole density is also important as it reduces the flow resistivity [5,7,8,19 ]. Although bottom-up approaches to fabricate such hexagonal hole arrays have previously been reported [20–23 ], no systematic studies into the limit of bottom-up fabrication and the RI sensor performances have been performed.
In this research, we demonstrate large-area fabrication of hexagonally ordered plasmonic nanohole arrays produced by a self-assembled colloid mask layer. We discuss the controllability of the nanohole size, using plasma etching of the colloid mask and compare the structure and optical properties of the fabricated hexagonal hole arrays. Furthermore, we carry out simulations to find the optimal hole size for RI sensing and field enhancement at both transmission dip and peak wavelengths in the optical spectra of the sensors. Since transmission dip wavelength is sensitive to the film surface and the peak wavelength is sensitive inside the nanoholes, comparison of sensing properties at these wavelengths is of vital importance for nanohole array based sensors [15,24 ].
2.1 Nanohole fabrication
The nanohole arrays were fabricated utilizing polystyrene (PS) nanosphere colloids self-assembled in a hexagonally close-packed manner, working as a mask layer on the substrate . The fabrication process is schematically shown in Fig. 1 . As substrates 1 × 1 cm2 glass slides were utilized (Matsunami, Japan). The glass slides where cleaned by successive rinsing in deionized water and ethanol, and were also subjected to a 4 min air plasma cleaning (Harrick Plasma PDC32G, USA). A 10 cm Petri dish used for the fabrication was cleaned using an UV–ozone cleaner. The PS sphere (diameter 320 nm) colloidal solution (8 wt.%, Life Technologies, USA) was diluted with deionized water to a concentration of 0.4 wt.%, and subsequently mixed with ethanol in a 1:1 ratio. The mixture was ultrasonicated for 20 min in order to disperse the PS colloids equally. The assist glass slide, which is thicker than the sample glass substrates, was put at the center of the Petri dish and the sample substrate placed nearby the assist slide . The Petri dish was placed in a desiccator and deionized water was added until we could observe the surface tension between the assist slide and the water. PS colloid solution (220 μl) was pipetted onto the surface of the assist slide, followed by gentle instillation of several droplets of sodium dodecyl sulfate (SDS) solution (~20 μl, 1 wt.%). This yields a close-packed PS sphere monolayer at the interface of air and liquid. Leaving the Petri dish in the desiccator and allowing the solution to slowly evaporate, the monolayer is transferred to the surface of the glass substrate.
After deposition, the diameter of the PS colloids can be controllably reduced by plasma etching (~10W, Harrik Plasma, USA). Two different pressure conditions during plasma etching were tested: i) One conducted at a constant pressure of 35 Pa by supplying air (constant pressure condition), and ii) the other without air supply causing the pressure decrease during the process as reactants were consumed and pumped out (no supply condition).
On substrates with size-reduced PS spheres masks, metal layers were deposited by sputtering. Sputtering pressure was 0.5 Pa in Ar gas atmosphere (back pressure: 1.0 x 10−5 Pa). The PS sphere monolayer acts as a mask, transferring the hexagonal pattern to the formed metal film. We used Au for the metal film (aimed to 23-25 nm), with a thin Al layer (1-2 nm) acting as an adhesion layer with the glass. As the last fabrication step, the PS spheres were removed from the substrate by tape stripping. This yields an Au film with a hexagonally ordered nanohole array.
The surface morphology of the achieved hexagonally close-packed PS colloids and nanohole arrays were characterized by a field-emission scanning electron microscope (SEM, JEOL-JSM 7000F, Japan) in secondary electron imaging mode. For observation of the PS colloid spheres on glass substrates, the acceleration voltage was set to 1 kV to avoid charging. For metallic nanohole arrays higher voltages (up to 20 kV) were applied for better spatial resolution. The diameters of the PS spheres after size-reduction and of the nanoholes were averaged for more than 55 data points. To evaluate film thickness distribution, an atomic force microscope (AFM, Hitachi, 5100N) was used.
The optical transmission spectra of the fabricated nanohole arrays were studied using a fiber-coupled spectrometer (Hamamatsu C10083MD, Japan) and fiber-coupled halogen lamp (Ocean Optics HL-2000, USA) connected to an optical microscope. A x50 objective was used with low illumination NA (<0.1) by defocusing the condenser lens. This yielded a circular detection area with an approximate diameter of 11 μm. The optical measurements were conducted in room temperature, and at ambient atmosphere. We changed the medium from air to water for refractive index (RI) sensitivity measurement. The sample was illuminated from the gold film side, except for the RI sensing measurement due to the liquid cell configuration. We note that the optical reciprocity well holds for such thin film samples and the illumination side has no influence in the transmission spectrum.
2.3 Electromagnetic simulation
Electromagnetic simulations were performed to extract the dependence of the optical properties and sensor performance purely on the nanohole size, avoiding possible fabrication errors as in the experiment. For this purpose a multiple multipole program (MMP) software package (OpenMaX) was used . MMP is a boundary method in frequency domain where the electric field is described as a sum of basis functions (so called “expansions”) that are the analytical solutions of Maxwell’s equation. A hexagonal lattice dimension corresponding to the experiment without the adhesion layer (only 25 nm thick Au layer) was set, and only the diameter of the hole was changed. To avoid infinite field enhancement, the edge of the hole was rounded by 8 nm. With hexagonal periodic boundary conditions, the efficiency of the Rayleigh expansion was evaluated to acquire transmission spectra with the normal plane wave illumination from the air side. Transmission peak and dip positions and intensities were determined by polynomial fitting. For RI sensitivity calculations, the medium was changed from air to water, and the resonance shift per unit RI is evaluated. We note that these sensitivity calculations by medium change from air to water tend to under-estimate the RI sensitivity compared to those measured at higher RI, since the sensitivity is roughly proportional to dielectric constant, which is the square of RI. However for calculation accuracy, we chose rather large medium RI change. The time-averaged electric field was extracted at the transmission peak and dip wavelengths. To obtain field enhancement factors the field values were normalized by the incident field. The field enhancement values at different positions were evaluated at 1 nm above the Au film surface along the film plane. We optimized the simulation model so that the relative average error was kept below 1.3%, typically in the range of 0.1% for most investigated structures. For dispersive gold dielectric constants, we used literature values . Constant RIs of 1.333 and 1.52 were used for water and glass respectively.
3.1 Self-assembly of colloids
By utilizing the process described above, we produced samples with hexagonally close-packed PS colloids. Figure 2(a) shows SEM images of the hexagonally ordered PS colloids (diameter 320 nm) before any etching. The achieved area of single domains was more than 20 × 20 μm2. However, also within the domains some line and point defects are observed. In order to study the influence of hole diameter on the plasmonic properties of nanohole arrays, the diameter of the PS colloids were reduced by plasma etching. SEM images of PS colloids etched for 4, 8, and 12 min are shown in Figs. 2(b)-2(d), respectively. It was observed that colloids located close to defects were etched faster than the fully surrounded ones. Therefore, with increased etching time, the uniformity of the PS monolayer is affected. We investigated the reduction of the PS sphere diameter by increasing the plasma etching time. The results of this study are summarized in Fig. 2(e). In both pressure conditions, no supply and constant pressure (explained in the method section), the diameter appeared to decrease more or less linearly but with different etching rates for the two conditions. An etching rate of 20.0 nm/min was observed for the constant pressure condition, and a rate of 7.72 nm/min for the no supply condition.
3.2 Hexagonal nanohole array
Hexagonally ordered Au nanohole arrays with varying diameters were fabricated as described above, utilizing the plasma etched PS colloid arrays as masks. SEM images of the fabricated nanohole arrays are shown in Fig. 3(a)-3(c) . From the SEM images, the average diameters of the nanoholes in the three arrays were determined to be 156 nm ( ± 9 nm), 210 nm ( ± 14 nm) and 238 nm ( ± 9 nm), produced from 156, 200, 230 nm colloid masks, respectively. As discussed above, when decreasing the diameter of nanoholes (increased etching time), the uniformity of the array is affected, as is evident from looking at the image of the smallest holes (Fig. 3(a)). We would also like the reader to note that if the colloids are etched to become too small, removal of the colloids by tape-stripping cannot be performed, as the colloid shape seems to lose its spherical shape, and without the under-cut the edge of the size-reduced colloids will be buried in the deposited metal layer. We found that the smallest achievable hole size was around half the size of the original sphere diameter. For the larger holes, it was difficult to control the film thickness because of the shadowing by the neighboring mask colloids during film deposition. The film deposition rate for larger holes is considerably reduced from that of a plane film without structure. The average film thickness of these three samples was controlled to be around 25 nm by adjusting the thickness for each colloid size. The achieved thickness was confirmed by AFM measurements. However, with larger colloid masks, it was difficult to achieve homogeneous thickness, which we discuss in the later section.
Figures 3(d)-3(f) show measured optical transmission spectra of the samples shown in the SEM images (Fig. 3(a)-3(c)). The transmission dip wavelengths are found to be 561, 569 and 631 nm; and transmission peaks 695, 692, 840 nm; for 156, 210 and 238 nm holes respectively. We note that the effective periodicity is constant at 277 nm for all the samples corresponding to the original colloid mask packing.
In the experiments, we found that the thickness control is difficult for larger colloid masks due to shadowing, and that the size control becomes difficult for smaller colloid masks due to the intrinsic inhomogeneity of the colloid and re-deposition during size reduction. However, it is still important to extract the error-free influence of the hole size on the optical properties and sensor performance in order to optimize the structural parameters. Simulation is helpful for such optimization purposes. Figure 4 shows simulated transmission spectra and RI sensitivity of ideal structures corresponding to those of the experiments. The summarized transmission peak and dip wavelengths in Fig. 4(b) indicate that for diameters smaller than 150 nm, the dip wavelength remains rather unchanged as the hole diameter is increased, whereas the peak wavelength shows a significant red-shift. For larger hole diameters, the transmission dip wavelength starts to blue-shift, which is related to the change of the effective surface plasmon wavelength by the presence of the hole . This effect should become more significant at larger hole diameters. Because of this blue-shifting of the dip, the transmission peak is also drawn towards shorter wavelengths. The wavelength difference between the dip and peak keeps increasing as the diameter increases, which relates to the polarizability of the hole itself [15, 28 ].
For RI sensing the optimal hole diameter for the highest sensitivity is somewhere around 160 nm, considering the sensitivities of both transmission peak and dip as shown in Fig. 4c. A hole diameter at around 160 nm also matches the highest intensity difference of the peak and dip, indicating the highest intensity contrast, or highest signal-to-noise sensor signal.
The electric field distributions at the transmission peak and dip wavelengths for representative hole diameters are shown in Fig. 5 . To emulate sensing in aqueous conditions, the medium was set to water. For all investigated systems, the field is distributed over the film at the transmission dip wavelengths while the field is localized inside the hole at the transmission peak wavelengths [15,24 ]. This can be more clearly seen in the line profile of the field enhancement in Fig. 6 . The field enhancement is highest at the edge of the hole for all diameters, which is natural as it is the sharpest edge. With increasing diameter the field inside the hole becomes weaker but the field outside grows stronger. For field enhancement the optimum diameter is at around 100 nm. Interestingly, this diameter corresponds to the highest RI sensitivity of the transmission dip, which is considered as a propagating mode, instead of that of the transmission peak [15,24 ].
4.1 Limitation of fabrication: comparison of experiment with simulation
The linear size reduction of colloid masks shown in Fig. 2(e) indicates that the size can be easily tuned. This is consistent with the report by Akinoglu et al., in which they propose a model of mixed isotropic and anisotropic etching processes . The etching rate is lower than their report since lower power plasma is used. In our experiments, the subsequent metal film deposition was, however, not so simple for larger hole sizes due to shadowing effects. The film thickness tends to be thinner at the thin gap between two colloids than at the larger triple-point area between three colloid spheres, as shown in the AFM images in Fig. 7 . This inhomogeneity seems to cause significant effects on the optical properties of the hole arrays. In the simulations the transmission dip for holes larger than 200 nm tends to blue-shift with increasing hole size, see Fig. 4. The superimposed experimental plots in Fig. 4(b) show fairly good agreement with the simulated results up to 220 nm hole diameter. The experimental RI sensitivity for 210 and 220 nm hole diameter, as shown in Fig. 4(c), negatively deviates from the simulation plots, which can be attributed to the error in the film thickness as well as the difference in the cross-sectional hole shape. The hole edge of the experimental structures are most likely less sharp compared to the simulated ones. For the larger hole (238nm), both transmission peak and dip significantly deviate from the experiment. (The transmission peak around 850 nm is out of range.) This is most probably related to the mentioned inhomogeneous thickness distribution due to the shadowing. Because of the same shadowing effect, at the defects in the array film thickness tend to be larger for large hole samples, also shown in Fig. 7(b). Since the excessive size reduction of colloid masks caused lift-off problems, the achievable smallest size was roughly half of the original colloid diameter.
Considering this together with the shadowing, the size range of holes that can be fabricated using this technique is in the range 50-70% of the original mask colloid diameter. This lower diameter limit is similar to the previous report for larger periodicity hole arrays . For large hole fabrication, it was also reported that too large PS sphere and thick film deposition could result in bridging of mask spheres and cause more inhomogeneity of the film thickness .
4.2 RI sensitivity and field enhancement
The nanohole array system is comprised of single hole oscillators coupled through Surface Plasmon Polariton (SPP) over the periodic lattice. This mixed system makes the optimization slightly more complicated than that for a single oscillator system. One also needs to consider the single hole together with the lattice. To better understand the highest field enhancement around the diameter of 100 nm, we compared the transmission peak and dip wavelengths with the resonance wavelength of a single hole, see Fig. 8 . The single hole resonance was calculated from the dispersion relation in water and an analytical formulation of a hole resonance condition [32,33 ]. At a diameter around 100 nm the single hole resonance and array resonance (transmission peak and dip) overlap at a wavelength of roughly 700 nm, which also matches the highest field enhancement. This can be relatively easily understood, as the polarizability is highest at this wavelength and the coupling to SPP is most efficient by matching the array resonance through SPP. We, however, have to note that this condition is typically out of the smallest hole diameter limit shown here by self-assembled hexagonal hole array (50% of original diameter), since the single nanohole resonance condition with effective half wavelength is shorter than 50% of SPP wavelength . Such engineering of field enhancement is important for e.g. surface-enhanced Raman spectroscopy (SERS) applications [34–36 ].
Optimization of RI sensitivity seems more complicated than that of the field enhancement as it also depends on how strongly SPP is excited, which is dependent on the polarizability of each hole. The polarizability of a sufficiently large hole at an off-resonant wavelength can exceed that of a smaller hole at resonance. In such a situation, which seems the case for our example system, the RI sensitivity of a large hole can be higher than a smaller one, although the field enhancement is stronger for the smaller hole at resonance.
As such, controlling both SPP coupling through periodic lattice and single hole resonance is important to optimize the sensor performance. The basic guideline is to match the single hole resonance and the lattice coupling. The controllable parameters are the film thickness, hole diameter, and lattice period, which effectively determines the SPP wavelength and single hole resonance.
Using self-assembly of colloids and size reduction by plasma etching, we have successfully fabricated hexagonally ordered plasmonic nanohole arrays with different hole diameters. Through this method, we found the range of the achievable hole sizes is from 50 to 70% of the diameter of the original colloid mask. The diameter can be well controlled by the plasma etching time, while the film thickness control turned out to be more difficult. The resonance wavelengths of properly fabricated nanohole arrays agreed well with those calculated from electromagnetic simulations. From simulation results, we also found that the optimal hole diameter for RI sensing with the highest signal-to-noise ratio is at a diameter around 160 nm, which can be attainable through the presented fabrication approach. For field enhancement, the optimum diameter is at around 100nm, which also coincides with the highest RI sensitivity of the transmission dip. At around this diameter, the resonance wavelengths of the single hole and lattice match at roughly 700 nm. However, this size is out of the capabilities of the presented fabrication method. Hence, further development of the method will be needed in order to reach this limit.
The knowledge of fabrication limitations and sensor design optimization are useful guidelines for nanhole based biosensor applications. Since periodic hole arrays can also work as grating wave coupler, integrated optical devices with biosensor elements could be realized using properly tuned nanohole arrays [37–39 ]. The fabrication limitations of the self-assembly based method may be pushed by e.g. shrinking the hole diameter through film deposition [40,41 ].
We acknowledge the financial support from Kazato Research Foundation and Kurata Memorial Hitachi Science and Technology Foundation, Murata Science Foundation, the Asahi Glass Foundation, the Shimadzu Science Foundation, and the Japan Society for the Promotion of Science #15F15744, 26870184, 25420707, 24108708.
References and links
3. N. C. Lindquist, P. Nagpal, K. M. McPeak, D. J. Norris, and S. H. Oh, “Engineering metallic nanostructures for plasmonics and nanophotonics,” Rep. Prog. Phys. 75(3), 036501 (2012). [CrossRef] [PubMed]
9. V. A. Zenin, A. Andryieuski, R. Malureanu, I. P. Radko, V. S. Volkov, D. K. Gramotnev, A. V. Lavrinenko, and S. I. Bozhevolnyi, “Boosting local field enhancement by on-chip nanofocusing and impedance-matched plasmonic antenna,” Nano Lett. 15(12), 8148–8154 (2015). [CrossRef] [PubMed]
10. M. P. Jonsson, A. B. Dahlin, P. Jönsson, and F. Höök, “Nanoplasmonic biosensing with focus on short-range ordered nanoholes in thin metal films,” Biointerphases 3(3), FD30–FD40 (2008). [CrossRef] [PubMed]
11. K. Nakamoto, R. Kurita, and O. Niwa, “Electrochemical surface plasmon resonance measurement based on gold nanohole array fabricated by nanoimprinting technique,” Anal. Chem. 84(7), 3187–3191 (2012). [CrossRef] [PubMed]
12. H. Im, H. Shao, Y. I. Park, V. M. Peterson, C. M. Castro, R. Weissleder, and H. Lee, “Label-free detection and molecular profiling of exosomes with a nano-plasmonic sensor,” Nat. Biotechnol. 32(5), 490–495 (2014). [CrossRef] [PubMed]
13. A. B. Dahlin, B. Dielacher, P. Rajendran, K. Sugihara, T. Sannomiya, M. Zenobi-Wong, and J. Vörös, “Electrochemical plasmonic sensors,” Anal. Bioanal. Chem. 402(5), 1773–1784 (2012). [CrossRef] [PubMed]
14. T. H. Reilly III, R. C. Tenent, T. M. Barnes, K. L. Rowlen, and J. van de Lagemaat, “Controlling the optical properties of plasmonic disordered nanohole silver films,” ACS Nano 4(2), 615–624 (2010). [CrossRef] [PubMed]
15. T. Sannomiya, O. Scholder, K. Jefimovs, C. Hafner, and A. B. Dahlin, “Investigation of Plasmon resonances in metal films with nanohole arrays for biosensing applications,” Small 7(12), 1653–1663 (2011). [CrossRef] [PubMed]
18. Y. Ekşioğlu, A. E. Cetin, and J. Petráček, “Optical response of plasmonic nanohole arrays: comparison of square and hexagonal lattices,” Plasmonics (2015), doi:. [CrossRef]
19. C. Escobedo, A. G. Brolo, R. Gordon, and D. Sinton, “Flow-Through vs Flow-Over: Analysis of Transport and Binding in Nanohole Array Plasmonic Biosensors,” Anal. Chem. 82(24), 10015–10020 (2010). [CrossRef] [PubMed]
20. A. Kosiorek, W. Kandulski, P. Chudzinski, K. Kempa, and M. Giersig, “Shadow nanosphere lithography: simulation and experiment,” Nano Lett. 4(7), 1359–1363 (2004). [CrossRef]
22. E. T. Papaioannou, V. Kapaklis, E. Melander, B. Hjörvarsson, S. D. Pappas, P. Patoka, M. Giersig, P. Fumagalli, A. Garcia-Martin, and G. Ctistis, “Surface plasmons and magneto-optic activity in hexagonal Ni anti-dot arrays,” Opt. Express 19(24), 23867–23877 (2011). [CrossRef] [PubMed]
23. H. W. Ting, Y. K. Lin, Y. J. Wu, L. J. Chou, C. J. Tsai, and L. J. Chen, “Large area controllable hexagonal close-packed single-crystalline metal nanocrystal arrays with localized surface plasmon resonance response,” J. Mater. Chem. C Mater. Opt. Electron. Devices 1(22), 3593–3599 (2013). [CrossRef]
24. M. Schwind, B. Kasemo, and I. Zorić, “Localized and Propagating plasmons in metal films with nanoholes,” Nano Lett. 13(4), 1743–1750 (2013). [PubMed]
26. C. Hafner, “Boundary methods for optical nano structures,” Phys. Status Solidi, B Basic Res. 244(10), 3435–3447 (2007). [CrossRef]
27. P. B. Johnson and R. W. Christy, “Optical constants of the noble metals,” Phys. Rev. B 6(12), 4370–4379 (1972). [CrossRef]
28. F. J. G. de Abajo, “Light scattering by particle and hole arrays,” Rev. Mod. Phys. 79(4), 1267–1290 (2007). [CrossRef]
29. E. M. Akinoglu, A. J. Morfa, and M. Giersig, “Understanding anisotropic plasma etching of two-dimensional polystyrene opals for advanced materials fabrication,” Langmuir 30(41), 12354–12361 (2014). [CrossRef] [PubMed]
30. M. P. Murray-Méthot, M. Ratel, and J. F. Masson, “Optical properties of Au, Ag, and bimetallic Au on Ag nanohole arrays,” J. Phys. Chem. C 114(18), 8268–8275 (2010). [CrossRef]
32. T. H. Park, N. Mirin, J. B. Lassiter, C. L. Nehl, N. J. Halas, and P. Nordlander, “Optical properties of a nanosized hole in a thin metallic film,” ACS Nano 2(1), 25–32 (2008). [CrossRef] [PubMed]
33. Y. Ikenoya, M. Susa, J. Shi, Y. Nakamura, A. B. Dahlin, and T. Sannomiya, “Optical resonances in short-range ordered nanoholes in ultrathin aluminum/aluminum nitride multilayers,” J. Phys. Chem. C 117(12), 6373–6382 (2013). [CrossRef]
34. A. Lesuffleur, H. Im, N. C. Lindquist, and S. H. Oh, “Periodic nanohole arrays with shape-enhanced plasmon resonance as real-time biosensors,” Appl. Phys. Lett. 90(24), 243110 (2007). [CrossRef]
35. T. Y. Jeon, S. G. Park, S. Y. Lee, H. C. Jeon, and S. M. Yang, “Shape control of Ag nanostructures for practical SERS substrates,” ACS Appl. Mater. Interfaces 5(2), 243–248 (2013). [CrossRef] [PubMed]
36. 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]
37. M. C. Estevez, M. Alvarez, and L. M. Lechuga, “Integrated optical devices for lab-on-a-chip biosensing applications,” Laser Photonics Rev. 6(4), 463–487 (2012). [CrossRef]
38. C. Ciminelli, C. M. Campanella, F. Dell’Olio, C. E. Campanella, and M. N. Armenise, “Label-free optical resonant sensors for biochemical applications,” Prog. Quantum Electron. 37(2), 51–107 (2013). [CrossRef]
39. F. Dell’Olio, D. Conteduca, C. Ciminelli, and M. N. Armenise, “New ultrasensitive resonant photonic platform for label-free biosensing,” Opt. Express 23(22), 28593–28604 (2015). [CrossRef] [PubMed]
40. S. Syrenova, C. Wadell, and C. Langhammer, “Shrinking-hole colloidal lithography: self-aligned nanofabrication of complex plasmonic nanoantennas,” Nano Lett. 14(5), 2655–2663 (2014). [CrossRef] [PubMed]
41. J. Junesch and T. Sannomiya, “Ultrathin suspended nanopores with surface plasmon resonance fabricated by combined colloidal lithography and film transfer,” ACS Appl. Mater. Interfaces 6(9), 6322–6331 (2014). [CrossRef] [PubMed]