We demonstrated the plasmonic metallic nanostructure fabricated by direct nanoimprinting of gold nanoparticles (AuNPs). This approach combines the patterning and lift-off processes into a simple one-step process without the need for expensive patterning lithographies and the stringent requirement of the lift-off process for nanostructures. Good imprinting integrity was accomplished with a negligible residual layer. The dynamic optical responses of the imprinted gold pillars from AuNPs to the bulk material during the annealing process were investigated. The localized surface plasmon resonance (LSPR) properties of AuNPs or gold pillar arrays can be controlled and tuned during the annealing process. The sensitivity of the gold pillar array in terms of the wavelength shift per refractive index unit (RIU) reached 259 nm/RIU. The size of the imprinted gold pillars is highly scalable in our process. The corresponding resonance wavelengths can be widely tuned from the visible to infrared region by changing the size of the gold pillars, thus providing a wide range of sensing capability.
© 2011 OSA
Metallic metamaterials have been demonstrated to support surface plasmon resonance (SPR) . There are two kinds of SPR. Propagating surface plasmon resonance  can be defined as the charge density wave propagating along the continuous metal-dielectric interface. Another kind of SPR is localized surface plasmon resonance (LSPR) [3, 4]. When light with a proper wavelength illuminates the metal structures, free electrons oscillate tremendously in a localized region. The electromagnetic field near metallic surface is highly enhanced. The spectral position of the plasmon resonance is sensitive to the dielectric environment within near field region. SPR-based sensors have been widely applied for biosensing owing to their benefits of high sensitivity, real-time monitoring, and label-free sample preparation [2, 5].
Traditionally, metallic nanostructures can be patterned by electron beam lithography (EBL)  or optical lithography  with subsequent metal evaporation and lift-off processes. Direct metallic deposition and patterning with focus ion beam (FIB) lithography is another approach to pattern metallic nanostructures . However, the aforementioned lithographies are expensive, and both EBL and FIB are time-consuming. Moreover, an undercut profile is required to facilitate the lift-off process. The process is difficult to control in a single-layer resist system, especially for lift-off nanostructures. Multilayer schemes were proposed to create undercut features [9, 10], and the whole process became even more complicated. Solution-processible gold nanoparticles (AuNPs) spin coated on the patterned resist were proposed [11, 12] to simplify the fabrication process without the use of a metal evaporation vacuum chamber. Nevertheless, an additional lift-off step was still required to remove the patterned resist. In contrast with the above-mentioned methods, bottom-up approaches, such as nanosphere lithography  and hole-mask colloidal lithography , provided simple and cost-efficient ways to pattern the metallic nanostructures. Although their self-assembling nature restricted the producible pattern shapes, several pattern shapes, such as nanodisc, triangular, nanoring, crescent, and nanocone, have been fabricated [13–16].
Recently, direct nanoimprinting of AuNPs was investigated to fabricate nano/microscale electronic devices [17, 18]. Nanoimprinting lithography (NIL) has attracted great attention as an alternative nanopatterning technology that allows the fabrication of two-dimensional (2D) or three-dimensional structures with nanoscale resolution . Compared to other lithographies, NIL has the advantages of being high throughput, low cost, and high resolution . With the direct nanoimprinting of AuNPs, this approach combines the patterning and lift-off processes into a simple one-step process. In previous works [17, 18], the electrical properties of nano/microscale electronic devices fabricated by this approach were studied. In contrast, we focused here on the optical behavior of the imprinted metallic nanostructures. 2D photonic crystals of gold pillars were fabricated by the direct nanoimprinting of AuNPs. The fabrication parameters were studied and optimized. The resonance behavior and the sensing property of the fabricated plasmonic nanostructures were demonstrated. The dynamic optical responses of the imprinted gold pillars from AuNPs to bulk material during the annealing process were investigated.
Figure 1 shows the overall process scheme for direct nanoimprinting of AuNPs. Our home-built imprinting platform with a compressed air press (CAP) is illustrated in Fig. 1(a), and the details are described in Ref . Polydimethylsiloxane (PDMS) based polymers were chosen as the working stamp materials due to their ability to absorb the solvent without deformation . PDMS is porous such that the solvent of the AuNPs can escape from the PDMS, which helps the AuNP solution to solidify, and the AuNP pattern can then be defined. The pattern of the 2D pillar array in a silicon master mold was defined by an electron-beam writer (Leica WEPRINT200) on an oxide layer. After the resist development, the pattern was transferred to the oxide layer by an RIE oxide etcher (TEL TE5000). The height of the pillar structure was controlled by the thickness of the oxide layer. Before transferring the pattern to the PDMS, an anti-sticking treatment was applied by vapor deposition of F13-TCS on the silicon master to avoid any possible sticking of the PDMS on the silicon . A scheme of the two-layer composite PDMS stamp was employed. The h-PDMS (hard PDMS) was used as a thin, stiff structural layer to increase the mechanical stability  to improve the pattern resolution and edge definition of the stamp. A s-PDMS (soft PDMS, Dow Corning Sylgard 184) was employed as a supporting slab to avoid stamp cracking. In fabricating the working stamp, a thin h-PDMS was coated on the silicon master and heated to 60°C for 30 minutes. Then s-PDMS was poured above the h-PDMS and was cured at 80°C for 60 minutes (Fig. 1(b)). After the curing process, the two-layer h/s-PDMS stamps could be easily torn off.
The AuNPs were synthesized using a two-phase reduction method and were encapsulated with a hexanethiol self-assembled monolayer (SAM), as described in Ref . Subsequently, 5% AuNPs were suspended in an α-terpineol carrier solvent. The as-synthesized nanoparticles were inspected by transmission electron microscopy (TEM). Based on the analysis of the TEM image by the software of SigmaScan Pro 5, their average size and the standard deviation of sizes were determined to be 2.31 nm and 0.433 nm, respectively, after counting 120 particles. The procedure of the direct nanoimprinting of the AuNPs is shown in Fig. 1(c)-(e). The 5% AuNP solution was dispensed on a substrate (silicon or glass), and the h/s-PDMS stamp was then placed above the solution. The stack of the stamp and substrate was placed in the imprint chamber. Then the CAP and the heating were applied for 20 minutes. After all the solvent evaporated, the h/s-PDMS stamp was demolded. The metallic nanostructures were further annealed at 250°C for 35 sec to fuse the encapsulated AuNPs together.
The imprinted samples were inspected using scanning electron microscopy (SEM, HITACHI S-4000). Their transmission spectra were measured using the UV-Visible-NIR spectrophotometer (JASCO V-670). The probe wavelength range was from 400 nm to 2500 nm. Simulations were performed to clarify the plasmonic behavior of the imprinted gold pillar array. The extinction spectra were calculated from the transmission spectra as follows:
The implemented rigorous coupled-wave analysis (RCWA) algorithm  and Mie analysis  were used to evaluate and verify the optical behavior of the samples. They were used as a basis for comparison with the experimental results.
3.1 Imprinting temperature
The viscosity of the imprinted material is an important experimental parameter in nanoimprinting . Because it is different from the traditional hot embossing process, no baking step was applied between the steps of spin coating and imprinting to remove the solvent in our process. Instead, the solvent, α-terpineol, acts as the medium to assist the AuNPs filling into the cavity of the imprinting stamp. The α-terpineol has a very wide viscosity range . When the heating temperature increases, its viscosity decreases gradually and provides a better filling capability during the imprinting process. On the other hand, if the heating temperature is too high, the rate of the solvent evaporating in the cavity is quicker than the rate of the vapor penetrating out of the PDMS stamp. The vapor accumulates in the cavity quickly, and the imprinted pattern could be damaged.
We compared three different heating temperatures at 60°C, 70°C, and 80°C under the applied pressure of 5 bar. Fig. 2 shows the SEM images of AuNPs imprinted on the silicon substrate with a negative h/s-PDMS stamp. The pattern of the stamp is a square hole array with a width of 400 nm and a pitch of 800 nm. The imprinting temperature of 70°C is found to have better imprinting integrity than those of 60°C and 80°C. We speculate that the solvent evaporation rate was at equilibrium with the rate of the vapor penetrating through the PDMS stamp around this temperature. At the imprinting temperature of 80°C, the solvent vapor accumulated in the cavity, and the top of the imprinted gold pillars was deformed, as shown in Fig. 2(c). On the other hand, when the heated temperature was at 60°C, the viscosity of the solvent was not low enough to allow the AuNPs to flow into the cavity. The residual Au layer was found as shown in Fig. 2(a). The mass of the AuNPs was only approximately 5% of the total AuNP solution such that the majority of the mass was from the α-terpineol. When the solvent was evaporated completely, the volume of the AuNPs was less than the original volume of the AuNP solution. Afterwards, the AuNPs were further annealed to fuse the nanoparticles into a bulk material. In this step, the Au-S bond was broken, and the mass of the AuNPs was reduced. Consequently, the actual size of the AuNPs was smaller than the size of the cavity.
3.2 Imprinting pressure
Another important parameter was the imprinting pressure. The residual layer thickness should be directly related to the imprinting pressure. The applied pressure should be large enough to force the AuNPs to flow into the stamp cavity. The optimum imprinting pressure was investigated by repeating the process using four imprint pressures. Fig. 3 shows the SEM images of the AuNPs imprinted on the silicon substrate at pressures from 3 bar to 6 bar. The imprinting temperature was kept at 70°C. The pattern of the stamp is a square hole array with a width of 800 nm and a pitch of 1.6 μm. Apparently, the residual Au layer decreased as the applied pressure increased. For example, when the applied pressure was 6 bar, the residual layer was minimal, but the patterned integrity was not ideal, as shown in Fig. 3(d). The volume of the cavity in the PDMS stamp was compressed at high pressure. For this reason, the solvent vapor and AuNPs competed significantly for space in the cavity. Therefore, the pattern was damaged. These defects may cause broadened LSPR spectra and reduce the resonance intensity. On the other hand, there was a small amount of residue while the imprinting pressure was 5 bar, as shown in Fig. 3(c). However, the residue was spread out in a small size with a low density, which should not influence the optical response of the periodically distributed gold pillar array. Hence, we selected the imprinting pressure at 5 bar for the optimal imprinting pressure.
3.3 Optical response
The LSPR properties of metallic structures can be tuned by varying the structure’s shape, size, composition, and dielectric environment [13, 29, 30]. Here we investigated the resonance property of the imprinted gold pillar array by changing the diameter of the gold array and its dielectric environment. The PDMS stamps we used were square hole arrays with pitches of 600 nm, 800 nm, and 1000 nm, respectively. All of the hole widths were half of the array pitches. We changed the substrate to glass for the following transmission measurement. After the direct imprinting and the final annealing processes, gold pillar arrays with diameters of 290 nm, 360 nm, and 415 nm were obtained. Their corresponding heights were 63 nm, 80 nm, and 79 nm, respectively. The diameters and heights of the gold pillar were obtained from the SEM inspections. The measured extinction spectra are illustrated by black curves, as shown in the top row of Fig. 4 . Their corresponding LSPR peaks were at 950 nm, 1204 nm, and 1502 nm, respectively. The LSPR peaks were modulated by the patterned size of the gold pillars. The corresponding simulated spectra by RCWA are illustrated by black curves in the bottom row of Fig. 4. The gold pillar was assumed to be a cylindrical pillar in the simulations. The geometrical parameters were the pillar’s diameter (d), pillar’s height (h), and the array’s pitch (p), as shown in the inset of Fig. 4(a). The measured and simulated spectra show good agreement.
We further changed the dielectric environment by coating 180-nm poly(methyl methacrylate) (PMMA) on the top of the gold pillar arrays. The measured extinction spectra are illustrated by red curves in the top row of Fig. 4. Their corresponding LSPR peaks were red-shifted to 1060 nm, 1297 nm, and 1620 nm. The sensitivity of the gold pillar array was defined as the LSPR wavelength shift per refractive index unit (RIU).
The refractive indices of PMMA are 1.4812, 1.4743, and 1.4562, respectively, at these resonance wavelengths . Therefore, the sensitivities of the gold pillar arrays were 229 nm/RIU, 196 nm/RIU, and 259 nm/RIU, respectively. The values are comparable to the plasmon-based sensors reported in the literatures [32–34]. The simulated spectra of the gold pillar arrays in the PMMA are shown by red curves in the bottom row of Fig. 4 for comparison. The positions of the simulated resonance peaks are consistent with those of the measured resonance peaks. There were additional short-wavelength peaks appearing after the coating of PMMA, which may be attributed to the LSPRs with the quadrupolar mode and even the octapolar mode . The higher order modes appeared significantly in the PMMA environment, which can also be confirmed from the RCWA simulations.
According to the work from Rubinstein’s group , the sensitivities of their gold island films can be higher when annealing at the temperature higher than 550°C. Their annealed individual islands were found to be single crystalline. However, in our process, the annealing was applied after removing the PDMS stamp and the pattern distortion may occur when annealing at such high temperature. Wang’s group reported that the index sensitivity increases as the apexes of Au nanoparticles get sharper . The moderate sensitivities of our gold pillar arrays may result from the rounding edges of the fabricated nanostructures. Further process optimization is under way to have a better control on the pattern morphology. Currently, we keep the annealing at lower temperature for the further applications relying on the low temperature process, such as the applications on the flexible substrates. The diameters of the imprinted gold pillars are highly scalable in our process. The corresponding resonance wavelengths can be widely tuned from the visible to the infrared region by changing the size of the gold pillars, providing a wide range of sensing capability.
3.4 Annealing effect
During the final annealing process, the imprinted AuNP pillar was fused into bulk gold at the heating temperature of 250°C. The measured extinction spectra of the AuNP pillar arrays heated for 15 sec, 25 sec, and 35 sec are shown in Fig. 5 . The unheated sample was also illustrated for comparison. The imprinted pillars were on glass substrate with a diameter of approximately 440 nm, a height of approximately 80 nm, and a pitch of 1200 nm. There was a resonance peak at the wavelength of 598 nm for the unheated sample. This is the LSPR originating from the AuNPs, which can be verified from the Mie scattering analysis of the AuNPs, as shown in Fig. 6(a) . In that analysis, the AuNPs were assumed to be ideal gold spheres with a diameter ranging from 3 nm to 200 nm. The dielectric environment was assumed to be air. The resonance peak was at the wavelength of 500 nm for the 20-nm AuNPs. The resonance peak was then red-shifted with the increasing size of the AuNPs. For the AuNPs larger than 100 nm, a high-order resonance peak appeared, and the resonance spectra became broad. Because the imprinted AuNPs were on the glass substrate and the shapes of the AuNPs may not be spherical, the measured LSPR peak coming from the AuNPs was at a longer wavelength compared to the simulated spectra.
By applying the annealing process at 250°C for 15 sec, the LSPR peak coming from the AuNPs was red-shifted, indicating that some of the AuNPs were fused into larger AuNPs, which can be explained from the Mie analysis shown in Fig. 6(a). The strength of the LSPR peak became smaller, indicating that the quantity of AuNPs should be reduced. On the other hand, in addition to the peak from the AuNPs, there was an LSPR peak arising around the wavelength of 1900 nm. This LSPR peak should come from the LSPR of the gold pillar array, which can be confirmed from the RCWA simulations described in previous section. The optical response of the pillars was found to be a combination of the individual AuNP LSPR and the LSPR due to the pillar array.
During the annealing process, the LSPR coming from the AuNPs became weaker, broader, and red-shifted. The LSPR coming from the gold pillar array became stronger, and its spectral bandwidth became narrower. Finally, the strongest LSPR peak at 1898 nm with an annealing treatment for 35 sec was obtained. The LSPR from the AuNPs disappeared. Most of the AuNPs should be fused to bulk gold at this time. The annealing can remove the effects of the individual AuNPs leaving only the LSPR due to the gold pillar arrays. The LSPR properties from the AuNPs or gold pillar arrays can be controlled and tuned during the annealing process.
To further investigate the plasmonic dependence on the imprinted AuNPs residual layer, the spin-coated 5% AuNPs on the glass substrate were annealed at 250°C directly without the imprinting process. Their extinction spectra varied with the annealing time from 0 sec to 45 sec were illustrated in Fig. 6(b). No LSPR peak was observed for the unheated sample, which is different from the case of the unheated imprinted sample shown in Fig. 5. This result could be due to two reasons: (1) the extinction peak of the small AuNPs with a diameter of approximately 3 nm was weak compared to the others, which can be confirmed from the simulation spectra in Fig. 6(a); and (2) the imprinted sample was heated at 70°C during the imprinting process, and some of AuNPs were fused to larger AuNPs in that stage. Therefore, there was a noticeable LSPR peak at 598 nm for the unannealed imprinted sample, as shown in Fig. 5.
Starting from the annealing time of 15 sec, the LSPR peak of the spin-coated AuNPs appeared at the wavelength of 584 nm as shown in Fig. 6(b). As the annealing time increased, the LSPR peaks of the AuNPs were red-shifted and located at the wavelengths of 596 nm, 636 nm, and 730 nm for the annealing time of 25 sec, 35 sec, and 45 sec, respectively. The extinction strength of the peak also increased with the annealing time. Both phenomena indicate that the AuNPs were fused into larger AuNPs with increasing annealing time, which is consistent with the trend of the simulation shown in Fig. 6(a). This phenomenon was due to the large surface energy of the AuNPs. They tend to aggregate and form larger AuNPs to lower their surface energy, and then the noticeable LSPR peak from the fused AuNPs appeared. For this reason, if a residual layer in the unpatterned region existed, then a LSPR peak should have arisen from the AuNPs in the residues. In our fabrication process, the optimum experimental parameters were tuned to have the least amount residues remaining, as described in previous sections. Therefore, the LSPR arising from the AuNPs was significantly suppressed, as illustrated in Fig. 4.
In this study, we synthesized AuNPs and nanoimprinted them into a imprinted plasmonic metallic nanostructure. This approach combines the patterning and lift-off processes into a simple one-step process without the need of expensive patterning lithographies and the stringent requirement of the lift-off process for nanostructures. The process conditions of the imprinting temperature and imprinting pressure were investigated and optimized. Good imprinting integrity was accomplished with a negligible residual layer, which was confirmed from the SEM inspections and the inexistence of the LSPR from the AuNPs. The mechanisms of the proper imprinting conditions were discussed. The dynamic optical responses of the imprinted gold pillars from AuNPs to bulk material during the annealing process were investigated. The LSPR properties from the AuNPs or gold pillar arrays can be controlled and tuned during the annealing process. The sensing performance of the 2D photonic crystals of the gold pillars was investigated with respect to the pillar’s diameter and their dielectric environment. The size of the imprinted gold pillars is highly scalable in our process. The corresponding resonance wavelengths can be widely tuned from the visible to infrared region by changing the size of the gold pillars, providing a wide range of sensing capability.
This study was supported by the National Science Council of Taiwan grants NSC 98-2221-E-006-018- and NSC 98-2218-E-009-001-.
References and Links
1. S. A. Maier, Plasmonics: Fundamentals and Applications (Springer, 2007).
2. J. Homola, S. S. Yee, and G. Gauglitz, “Surface plasmon resonance sensors: review,” Sens. Actuators B Chem. 54(1–2), 3–15 (1999). [CrossRef]
3. E. Hutter and J. H. Fendler, “Exploitation of localized surface plasmon resonance,” Adv. Mater. 16(19), 1685–1706 (2004). [CrossRef]
6. S. Gorelick, V. A. Guzenko, J. Vila-Comamala, and C. David, “Direct e-beam writing of dense and high aspect ratio nanostructures in thick layers of PMMA for electroplating,” Nanotechnology 21(29), 295303 (2010). [CrossRef] [PubMed]
7. H. H. Solak, C. David, J. Gobrecht, V. Golovkina, F. Cerrina, S. O. Kim, and P. F. Nealey, “Sub-50 nm period patterns with EUV interference lithography,” Microelectron. Eng. 67–68, 56–62 (2003). [CrossRef]
8. S. Ahn, S. Kim, and H. Jeon, “Single-defect photonic crystal cavity laser fabricated by a combination of laser holography and focused ion beam lithography,” Appl. Phys. Lett. 96(13), 131101 (2010). [CrossRef]
9. G.-Y. Jung, E. Johnston-Halperin, W. Wu, Z. Yu, S.-Y. Wang, W. M. Tong, Z. Li, J. E. Green, B. A. Sheriff, A. Boukai, Y. Bunimovich, J. R. Heath, and R. S. Williams, “Circuit fabrication at 17 nm half-pitch by nanoimprint lithography,” Nano Lett. 6(3), 351–354 (2006). [CrossRef] [PubMed]
10. J. Wan, Z. Shu, S.-R. Deng, S.-Q. Xie, B.-R. Lu, R. Liu, Y. Chen, and X.-P. Qu, “Duplication of nanoimprint templates by a novel SU-8/SiO[sub 2]/PMMA trilayer technique,” J. Vac. Sci. Technol. B 27(1), 19–22 (2009). [CrossRef]
13. C. L. Haynes and R. P. Van Duyne, “Nanosphere Lithography: A Versatile Nanofabrication Tool for Studies of Size-Dependent Nanoparticle Optics,” J. Phys. Chem. B 105(24), 5599–5611 (2001). [CrossRef]
14. H. Fredriksson, Y. Alaverdyan, A. Dmitriev, C. Langhammer, D. S. Sutherland, M. Zäch, and B. Kasemo, “Hole–Mask Colloidal Lithography,” Adv. Mater. 19(23), 4297–4302 (2007). [CrossRef]
16. J. S. Shumaker-Parry, H. Rochholz, and M. Kreiter, “Fabrication of Crescent-Shaped Optical Antennas,” Adv. Mater. 17(17), 2131–2134 (2005). [CrossRef]
17. S. H. Ko, I. Park, H. Pan, C. P. Grigoropoulos, A. P. Pisano, C. K. Luscombe, and J. M. J. Fréchet, “Direct nanoimprinting of metal nanoparticles for nanoscale electronics fabrication,” Nano Lett. 7(7), 1869–1877 (2007). [CrossRef] [PubMed]
18. I. Park, S. H. Ko, H. Pan, C. P. Grigoropoulos, A. P. Pisano, J. M. J. Frechet, E. S. Lee, and J. H. Jeong, “Nanoscale patterning and electronics on flexible substrate by direct nanoimprinting of metallic nanoparticles,” Adv. Mater. 20(3), 489–496 (2008). [CrossRef]
19. S. Y. Chou, P. R. Krauss, and P. J. Renstrom, “Imprint lithography with 25-nanometer resolution,” Science 272(5258), 85–87 (1996). [CrossRef]
20. L. J. Guo, “Nanoimprint lithography: Methods and material requirements,” Adv. Mater. 19(4), 495–513 (2007). [CrossRef]
21. C.-H. Lin, H.-H. Lin, W.-Y. Chen, and T.-C. Cheng, “Direct imprinting on a polycarbonate substrate with a compressed air press for polarizer applications,” Microelectronic Engineering, (http://dx.doi.org/10.1016/j.mee.2010.1012.1089) (2011).
23. M. Beck, M. Graczyk, I. Maximov, E. L. Sarwe, T. G. I. Ling, M. Keil, and L. Montelius, “Improving stamps for 10 nm level wafer scale nanoimprint lithography,” Microelectron. Eng. 61–62(1–3), 441–448 (2002). [CrossRef]
24. T. W. Odom, J. C. Love, D. B. Wolfe, K. E. Paul, and G. M. Whitesides, “Improved pattern transfer in soft lithography using composite stamps,” Langmuir 18(13), 5314–5320 (2002). [CrossRef]
25. M. J. Hostetler, J. E. Wingate, C. J. Zhong, J. E. Harris, R. W. Vachet, M. R. Clark, J. D. Londono, S. J. Green, J. J. Stokes, G. D. Wignall, G. L. Glish, M. D. Porter, N. D. Evans, and R. W. Murray, “Alkanethiolate gold cluster molecules with core diameters from 1.5 to 5.2 nm: Core and monolayer properties as a function of core size,” Langmuir 14(1), 17–30 (1998). [CrossRef]
26. C.-H. Lin, H.-L. Chen, W.-C. Chao, C.-I. Hsieh, and W.-H. Chang, “Optical characterization of two-dimensional photonic crystals based on spectroscopic ellipsometry with rigorous coupled-wave analysis,” Microelectron. Eng. 83(4–9), 1798–1804 (2006). [CrossRef]
27. C. F. Bohren, and D. R. Huffman, “Absorption and Scattering by a Sphere,” in Absorption and Scattering of Light by Small Particles (John Wiley, New York, 1983), pp. 82–129.
28. C. M. S. Torres, “Nanostructure Science and Technology,” in Alternative Lithography: Unleashing The Potentials Of Nanotechnology D. J. Lockwood, ed. (Kluwer Academic, Plenum Germany, 2003).
31. S. N. Kasarova, N. G. Sultanova, C. D. Ivanov, and I. D. Nikolov, “Analysis of the dispersion of optical plastic materials,” Opt. Mater. 29(11), 1481–1490 (2007). [CrossRef]
32. K. M. Mayer, S. Lee, H. Liao, B. C. Rostro, A. Fuentes, P. T. Scully, C. L. Nehl, and J. H. Hafner, “A label-free immunoassay based upon localized surface plasmon resonance of gold nanorods,” ACS Nano 2(4), 687–692 (2008). [CrossRef]
34. J. Henzie, M. H. Lee, and T. W. Odom, “Multiscale patterning of plasmonic metamaterials,” Nat. Nanotechnol. 2(9), 549–554 (2007). [CrossRef]
35. C. Langhammer, B. Kasemo, and I. Zoric, “Absorption and scattering of light by Pt, Pd, Ag, and Au nanodisks: absolute cross sections and branching ratios,” J. Chem. Phys. 126(19), 194702 (2007). [CrossRef] [PubMed]
36. T. Karakouz, D. Holder, M. Goomanovsky, A. Vaskevich, and I. Rubinstein, “Morphology and Refractive Index Sensitivity of Gold Island Films,” Chem. Mater. 21(24), 5875–5885 (2009). [CrossRef]