Strong Coupling between Plasmonic Gap Modes and Photonic Lattice

Jun 5, 2015 - †Department of Materials Science and Engineering, ‡International Institute for Nanotechnology, §Department of Electrical Engineerin...
0 downloads 15 Views 2MB Size
Letter pubs.acs.org/NanoLett

Strong Coupling between Plasmonic Gap Modes and Photonic Lattice Modes in DNA-Assembled Gold Nanocube Arrays Qing-Yuan Lin,†,‡ Zhongyang Li,‡,§ Keith A. Brown,‡,∥ Matthew N. O’Brien,‡,∥ Michael B. Ross,‡,∥ Yu Zhou,†,‡ Serkan Butun,‡,§ Peng-Cheng Chen,†,‡ George C. Schatz,‡,∥ Vinayak P. Dravid,†,‡ Koray Aydin,*,‡,§ and Chad A. Mirkin*,†,‡,∥ †

Department of Materials Science and Engineering, ‡International Institute for Nanotechnology, §Department of Electrical Engineering and Computer Science, and ∥Department of Chemistry, Northwestern University, 2145 Sheridan Road, Evanston, Illinois 60208, United States S Supporting Information *

ABSTRACT: Control of both photonic and plasmonic coupling in a single optical device represents a challenge due to the distinct length scales that must be manipulated. Here, we show that optical metasurfaces with such control can be constructed using an approach that combines top-down and bottom-up processes, wherein gold nanocubes are assembled into ordered arrays via DNA hybridization events onto a gold film decorated with DNA-binding regions defined using electron beam lithography. This approach enables one to systematically tune three critical architectural parameters: (1) anisotropic metal nanoparticle shape and size, (2) the distance between nanoparticles and a metal surface, and (3) the symmetry and spacing of particles. Importantly, these parameters allow for the independent control of two distinct optical modes, a gap mode between the particle and the surface and a lattice mode that originates from cooperative scattering of many particles in an array. Through reflectivity spectroscopy and finite-difference timedomain simulation, we find that these modes can be brought into resonance and coupled strongly. The high degree of synthetic control enables the systematic study of this coupling with respect to geometry, lattice symmetry, and particle shape, which together serve as a compelling example of how nanoparticle-based optics can be useful to realize advanced nanophotonic structures that hold implications for sensing, quantum plasmonics, and tunable absorbers. KEYWORDS: Noble metal nanoparticles, gold nanocubes, DNA-mediated assembly, gap mode, lattice mode, plasmonic, photonic

T

photonic crystal community has long recognized that highly dispersive optical properties can be realized with arrays of materials with periodicity commensurate with the wavelength of light.17 In recent years, the plasmonics community has extended this concept by using top-down lithography techniques to construct periodic plasmonic nanostructures with sharp resonance features18−20 to study effects such as lasing,21 perfect absorption,22 light concentration and manipulation,23 and SERS.24 Many of these structures preserve the high field enhancements25,26 (critical for applications like SERS and lasing) inherent to plasmonics and dramatically increase their quality factor.27 However, top-down techniques such as physical vapor deposition produce polycrystalline metal structures, where the structural defects negatively impact device performance,28,29 and are limited to essentially planar architectures. As a result, various methods have been explored to merge the strengths of top-down and bottom-up approaches, wherein single crystalline nanoparticles can be assembled onto

he optical properties of noble metal nanostructures are dictated by the collective electric-field driven response of their conduction band electrons, a phenomenon known as a localized surface plasmon resonance (LSPR).1 The resonance conditions are dependent upon the nanostructure composition, shape, and size,2−4 and these parameters can be easily tuned via chemical synthesis.5−7 Synthetic control over nanoparticle plasmonic properties has enabled their use in a diverse set of applications including sensing,8 nonlinear optics,9 and surfaceenhanced Raman spectroscopy (SERS).10 A common strategy for enhancing the local electromagnetic field generated by such colloidal nanoparticles irradiated at their LSPR is to use bottom up assembly approaches to position them close (relative to their size) to other metal structures such that their plasmonic resonances hybridize.11−13 For example, when silver cubes are positioned at a fixed distance above a metal film, a plasmonic “gap mode” emerges that allows for tunable absorption and enhanced fluorescence emission.14−16 While control of the local interparticle geometry at the nanometer scale offers a promising route to access unique near-field optical responses, long-range order on the optical wavelength scale can be used to tune farfield interactions between particles and light. In particular, the © XXXX American Chemical Society

Received: April 21, 2015

A

DOI: 10.1021/acs.nanolett.5b01548 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters patterned substrates with programmed capture sites.30−33 In particular, it has recently been shown that complementary interactions between DNA-functionalized nanoparticles and a DNA-functionalized substrate can be used to assemble nanoparticles with control over their position and distance from the surface.34,35 Despite these synthetic advances, examples of colloidal nanoparticle-based optical devices consist of either randomly assembled particles that lack long-range order14 or lattices of nanoparticles spaced too closely to couple by coherent diffractive scattering of visible light,36 and thus the implications of controlling both photonic and plasmonic coupling in these devices requires further exploration. Here, we report the realization of strong coupling between a plasmonic gap mode and a photonic lattice mode in twodimensional lattices of gold nanocubes on gold substrates assembled with DNA. In particular, gold cubes positioned a fixed distance above a gold substrate (Figure 1a) represent an

excellent test system in which to study plasmonic and photonic coupling because isolated cubes on substrates have been shown to support a tunable cavity-like gap mode between the cube and film.37 This behavior can be demonstrated by computing the intensity distribution of the electric field (E-field) using finitedifference time-domain (FDTD) simulations (Figure 1b). However, new modes are predicted to arise if many such cubes are arranged periodically due to their collective interaction with incident light. Specifically, when arranged in a square lattice above a reflective substrate (Figure 1c), the outof-plane dipoles of these particles couple to support a collective surface lattice resonance (SLR) mode that is analogous to a Bloch mode (Figure 1d).38 In principle, it should be possible to independently tune these modes by changing the geometry of the system and bring these modes into resonance, a condition that has yielded strong field enhancement from the interaction of photonic and plasmonic modes in analogous systems.25 In a typical experiment to explore these emergent optical modes, gold nanocubes were assembled in a DNA-mediated fashion into arrays (Figure 2a). Electron beam lithography

Figure 2. Assembly and SEM characterization of a cube array. (a) Construction of a gold cube array on gold substrate. A silicon wafer coated with 100 nm of gold was spin coated with 50 nm of PMMA and EBL was carried out to define an array of trenches. Subsequently, the exposed gold surfaces were functionalized with DNA and cubes were immobilized into the trenches using complementary DNA hybridization. (b) SEM image of a typical cube array. The image was taken at backscattered mode, so the PMMA is invisible. Inset: a zoomed-in SEM image showing only 4 gold cubes (inset scale bar 300 nm).

(EBL) was used to pattern poly(methyl methacrylate) (PMMA) trenches in uniform arrays on Au-coated Si substrates over 200 × 200 um2 areas, where the bottom of each trench consisted of exposed Au (Supporting Information, Figure S2). The trench size and shape were matched to the nanoparticle species of interest (single crystalline gold cubes with average edge lengths of 73 ± 4 nm synthesized via a seed-mediated process7) such that only a single cube was captured in each trench. Cubes and the exposed gold in the trenches were each densely functionalized with a different thiolated DNA sequence (Supporting Information, Table S1) and hybridized with a complementary oligonucleotide to produce a rigid, doublestranded DNA shell with a short single-stranded region at the terminus. Cubes were then assembled into the trenches by designing the single-stranded region on the surface and cube to have complementary DNA sequences (Figure 2b, Supporting

Figure 1. Schematics and FDTD simulations showing the formation of gap and lattice modes. (a) Schematic drawing of a DNA-functionalized gold cube assembled onto a gold substrate through DNA hybridization, forming a programmable gap between the cube and the metal surface. DNA on cube (blue) and DNA on substrate (red) are designed to be complementary so that they will hybridize to facilitate the assembly of the cube. (b) Calculated E-field intensity profile for a 73 nm edge length cube separated from a gold film by 4.6 nm when illuminated by light at 745 nm. This wavelength corresponds to the gap mode as evident by the high field intensity in the gap. (c) Schematic drawing of an array of DNA-assembled gold cubes. (d) Calculated E-field intensity profile for the same structure as in (c) but illuminated at 837 nm. The standing waves with high intensity between the cubes indicate that this wavelength corresponds to the lattice mode of this array. B

DOI: 10.1021/acs.nanolett.5b01548 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Figure 3. Strong coupling between lattice and gap modes. (a) Reflectivity spectra of cube arrays of different periodicity P. From bottom to top, P increases from 500 to 950 nm in 50 nm increments. (b) FDTD simulation of reflectivity spectra of nanocube arrays with P from 500 to 1000 nm. Overlaid red squares, green triangles, blue circles represent the experimentally measured locations of the first order lattice modes, second order lattice modes, and gap modes, respectively. The color bar represents computed reflectivity intensity. (c) Plots of the resonance intensity evolution of the three modes shown in (b).

recently reported mode splitting of 3D plasmonic photonic crystals realized through DNA programmed assembly.36 As P is increased further (Figure 3a, red through black lines), a secondorder lattice mode was also observed, leading to a second anticrossing. We attribute the breadth of the SLR peak (relative to what has been observed in systems of plasmonic particles defined exclusively by top down lithography) to slight variations in particle size, position, and the relatively small size of the array compared to prior work, all properties that can ultimately be optimized in subsequent investigations.20 To further confirm our assignment of the observed modes, 3D FDTD simulations were performed using commercially available software (Lumerical, see Supporting Information for more details), allowing us to calculate reflectivity for nanocube arrays versus wavelength and P (Figure 3b). An anticrossing between the lattice and gap modes was also observed, which is consistent with experiments. Remarkably, the experimentally determined spectral location of the modes agree extremely well with the calculated spectra, as shown by the squares, triangles, and circles that correspond to the spectral positions of the first lattice modes, second lattice modes, and gap modes, respectively. Furthermore, the intensities of the modes, as defined by reflectivity resonance amplitude, exhibited a very clear transition of resonance intensities between lattice modes and gap modes (Figure 3c). Interestingly, one would expect a decrease in the lattice mode intensity as P was increased based solely upon the decrease in particle density. However, in the anticrossing region the lattice mode intensity (squares) increased dramatically while the gap mode intensity (circles) was considerably reduced, indicating that the modal intensities are dependent upon the degree to which they couple, as expected. To further explore the origins of the lattice and gap modes, we performed additional experiments investigating the role of nanoparticle shape and long-range order. Specifically, we compared three systems with the same particle density, but different experimental parameters: (1) cubes arranged in a

Information Figure S3). Importantly, SEM characterization of over 5000 trenches revealed that >98% of them were occupied by cubes with minimal nonspecific nanoparticle adsorption on the PMMA. After assembly, the substrate was dried in air, which results in a 4.6 nm gap between the particles and substrates, as previously measured by grazing incidence smallangle X-ray scattering (GISAXS) for the same DNA design.39 The resonance wavelength of the gap mode can be modulated through manipulation of the particle−substrate gap distance,37 which can be programmed based on the DNA length.39 In order to experimentally characterize the interaction between the gap and lattice modes, optical measurements were performed using an inverted optical microscope in reflection mode. In a typical experiment, reflectivity spectra were collected in the wavelength range between 600 and 900 nm and normalized using reflectance measurements of a 100 nm thick gold film, which is the same film used to support the nanoparticles. To explore the consequences of spectral overlap between the gap mode and lattice mode, cube arrays with periodicities (P) of 500−950 nm, in 50 nm increments, were measured (Figure 3a). Beginning at the densest arrays (Figure 3a, purple line), two dips in reflectivity were observed: a longer wavelength dip corresponding to the gap mode and a shorter wavelength peak corresponding to the lattice mode. As P was increased (Figure 3a, dark blue line), the longer wavelength mode was spectrally stable while the shorter wavelength mode redshifted, in support of our mode assignment as the gap and lattice modes, respectively. Interestingly, as P was further increased (Figure 3a, light blue, green, light green, and orange lines), an anticrossing between the modes was observed, indicating that near P = 650 nm the modes were hybridized and photons with wavelengths near 750 nm exist in a superposition of these two modes. Specifically, this anticrossing has a Rabi splitting of ∼110 meV, which is calculated from the energy difference between the two modes at the resonant coupling point.40 This anticrossing suggests a strong coupling between gap modes and lattice modes41,42 and is comparable to a C

DOI: 10.1021/acs.nanolett.5b01548 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Figure 4. Experimental and computational study of an ordered nanocube array (left column), disordered nanocube array (middle column), and an ordered nanosphere array (right column). (a−c) SEM images for three constructed structures. (d−f) Calculated (solid lines) and measured (dotted lines) reflectivity spectra. (g−i) E-field spatial distributions computed using FDTD at incident light wavelength of 620 nm, as indicated by red dashed lines in (d−f). (j−l) E-field spatial distributions at 745 nm, as indicated by green dashed lines in (d−f). Scale bars, 500 nm.

square lattice with P = 500 nm (Figure 4a), shown above to exhibit both a lattice and a gap mode, (2) cubes randomly positioned (Figure 4b), which we would expect to exhibit a gap mode but no lattice mode, and (3) spheres positioned with the same periodicity as 1 (Figure 4c), which we would expect to exhibit a lattice mode but minimal gap mode. For an ordered array of cubes, two distinct minima were observed in both the experimental (Figure 4d, dotted line) and computed (Figure 4d, solid line) reflectivity spectra. As discussed above, the lower and higher wavelength modes correspond to the lattice mode and gap mode, as evident by the E-field distribution (Figure 4g,j, respectively). When cubes with and without long-range order are compared, the lattice mode disappeared for the randomly positioned sample, while the gap mode persisted (Figure 4e, dotted line), as expected. This experimental observation matches simulated spectra of an isolated particle, which lacks photonic coupling between nanoparticles (Figure 4e, solid line). Additionally, the E-field profiles exhibited no lattice mode for the randomly positioned cube sample (Figure 4h), while the gap mode was still present (Figure 4k). When periodic arrays of spheres were compared to those of cubes, neither lattice nor gap modes were observed in the simulated or experimental reflectivity spectra (Figure 4f). We hypothesize that the lack of a gap mode originates from the distinctly larger average separation between the surface of a sphere and a flat substrate relative to a cube. This geometric difference is manifested as a dramatically different E-field profile for surfacebound cubes (Figure 4j,k) and spheres (Figure 4l). The lack of a lattice mode for spheres (Figure 4i) is indicative of their weak out-of-plane dipole moment as compared with cubes, as shown previously for other particle structures.43

In summary, we have shown that DNA-mediated nanoparticle assembly can be used to construct optical metasurfaces with independent control of gap and lattice modes. Importantly, this synthetic control enables one to systematically tune the degree of coupling between these two modes to elucidate structure−property relationships. Importantly, this work highlights the potential for DNA-mediated assembly to be used for the study of nanophotonic properties and discovery of novel optical phenomena.44 When combined with large area nanopatterning tools such as beam pen lithography45,46 nanoimprint lithography,47 or molecular printing techniques such as dip pen nanolithography 48 or polymer pen lithography,49 this approach opens new opportunities for realizing large scale metasurfaces that consist of high quality plasmonic units and would be useful in applications such as lasing,50 fluorescence,51 and SERS.24



ASSOCIATED CONTENT

S Supporting Information *

Experimental details for nanoparticle synthesis, substrate fabrication, DNA synthesis, purification and functionalization of particles and substrates, nanocube array assembly, optical measurements, and FDTD simulations. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.nanolett.5b01548.



AUTHOR INFORMATION

Corresponding Authors

*(K.A.) E-mail: [email protected]. *(C.A.M.) E-mail: [email protected]. Author Contributions

Q.-Y.L. and Z.L. contributed equally to this work D

DOI: 10.1021/acs.nanolett.5b01548 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters Notes

(21) Zhou, W.; Dridi, M.; Suh, J. Y.; Kim, C. H.; Co, D. T.; Wasielewski, M. R.; Schatz, G. C.; Odom, T. W. Nat. Nanotechnol. 2013, 8, 506−511. (22) Yi, F.; Shim, E.; Zhu, A. Y.; Zhu, H.; Reed, J. C.; Cubukcu, E. Appl. Phys. Lett. 2013, 102, 221102. (23) Schuller, J. A.; Barnard, E. S.; Cai, W.; Jun, Y. C.; White, J. S.; Brongersma, M. L. Nat. Mater. 2010, 9, 193−204. (24) Haes, A. J.; Haynes, C. L.; McFarland, A. D.; Schatz, G. C.; Van Duyne, R. P.; Zou, S. MRS Bull. 2005, 30, 368−375. (25) Zou, S.; Janel, N.; Schatz, G. C. J. Chem. Phys. 2004, 120, 10871−10875. (26) Lassiter, J. B.; Chen, X.; Liu, X.; Ciracì, C.; Hoang, T. B.; Larouche, S.; Oh, S.-H.; Mikkelsen, M. H.; Smith, D. R. ACS Photonics 2014, 1, 1212−1217. (27) Li, Z.; Butun, S.; Aydin, K. ACS Photonics 2014, 1, 228−234. (28) Tang, Y.; Ouyang, M. Nat. Mater. 2007, 6, 754−759. (29) Goubet, N.; Yan, C.; Polli, D.; Portalès, H.; Arfaoui, I.; Cerullo, G.; Pileni, M.-P. Nano Lett. 2013, 13, 504−508. (30) Yan, B.; Thubagere, A.; Premasiri, W. R.; Ziegler, L. D.; Dal Negro, L.; Reinhard, B. M. ACS Nano 2009, 3, 1190−1202. (31) Henzie, J.; Andrews, S. C.; Ling, X. Y.; Li, Z.; Yang, P. Proc. Natl. Acad. Sci. U.S.A. 2013, 110, 6640−6645. (32) Zhou, X.; Zhou, Y.; Ku, J. C.; Zhang, C.; Mirkin, C. A. ACS Nano 2014, 8, 1511−1516. (33) Zhou, Y.; Zhou, X.; Park, D. J.; Torabi, K.; Brown, K. A.; Jones, M. R.; Zhang, C.; Schatz, G. C.; Mirkin, C. A. Nano Lett. 2014, 14, 2157−2161. (34) Hellstrom, S. L.; Kim, Y.; Fakonas, J. S.; Senesi, A. J.; Macfarlane, R. J.; Mirkin, C. A.; Atwater, H. A. Nano Lett. 2013, 13, 6084−6090. (35) O’Brien, M. N.; Radha, B.; Brown, K. A.; Jones, M. R.; Mirkin, C. A. Angew. Chem., Int. Ed. 2014, 53, 9532−9538. (36) Park, D. J.; Zhang, C.; Ku, J. C.; Zhou, Y.; Schatz, G. C.; Mirkin, C. A. Proc. Natl. Acad. Sci. U.S.A. 2015, 112, 977−981. (37) Lassiter, J. B.; McGuire, F.; Mock, J. J.; Ciracì, C.; Hill, R. T.; Wiley, B. J.; Chilkoti, A.; Smith, D. R. Nano Lett. 2013, 13, 5866− 5872. (38) Li, Z.; Butun, S.; Aydin, K. ACS Nano 2014, 8, 8242−8248. (39) Radha, B.; Senesi, A. J.; O’Brien, M. N.; Wang, M. X.; Auyeung, E.; Lee, B.; Mirkin, C. A. Nano Lett. 2014, 14, 2162−2167. (40) Christ, A.; Tikhodeev, S. G.; Gippius, N. A.; Kuhl, J.; Giessen, H. Phys. Rev. Lett. 2003, 91, 183901. (41) Yoshie, T.; Scherer, A.; Hendrickson, J.; Khitrova, G.; Gibbs, H. M.; Rupper, G.; Ell, C.; Shchekin, O. B.; Deppe, D. G. Nature 2004, 432, 200−203. (42) Zhou, W.; Suh, J. Y.; Hua, Y.; Odom, T. W. J. Phys. Chem. C 2013, 117, 2541−2546. (43) Zou, S.; Schatz, G. C. J. Chem. Phys. 2004, 121, 12606−12612. (44) Ross, M. B.; Blaber, M. G.; Schatz, G. C. Nat. Commun. 2014, 5, 4090. (45) Liao, X.; Brown, K. A.; Schmucker, A. L.; Liu, G.; He, S.; Shim, W.; Mirkin, C. A. Nat. Commun. 2013, 4, 2103. (46) Zhou, Y.; Xie, Z.; Brown, K. A.; Park, D. J.; Zhou, X.; Chen, P.C.; Hirtz, M.; Lin, Q.-Y.; Dravid, V. P.; Schatz, G. C.; Zheng, Z.; Mirkin, C. A. Small 2014, 11, 913−918. (47) Choi, S.-J.; Yoo, P. J.; Baek, S. J.; Kim, T. W.; Lee, H. H. J. Am. Chem. Soc. 2004, 126, 7744−7745. (48) Piner, R. D.; Zhu, J.; Xu, F.; Hong, S.; Mirkin, C. A. Science 1999, 283, 661−663. (49) Huo, F.; Zheng, Z.; Zheng, G.; Giam, L. R.; Zhang, H.; Mirkin, C. A. Science 2008, 321, 1658−1660. (50) van Beijnum, F.; van Veldhoven, P. J.; Geluk, E. J.; de Dood, M. J. A.; ’t Hooft, G. W.; van Exter, M. P. Phys. Rev. Lett. 2013, 110, 206802. (51) Laroche, M.; Albaladejo, S.; Carminati, R.; Sáenz, J. J. Opt. Lett. 2007, 32, 2762−2764.

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This material is based upon work supported by the AFOSR awards FA9550-12-1-0280 and FA9550-11-1-0275 and the National Science Foundation’s MRSEC program (DMR1121262) at the Materials Research Center of Northwestern University. Q.-Y.L. acknowledges support from the Hierarchical Materials Cluster Program Fellowship from Northwestern University. Z.L. and Y.Z. gratefully acknowledges support from the Ryan Fellowship at Northwestern University. K.A.B. acknowledges support from Northwestern University’s International Institute for Nanotechnology. M.N.O. acknowledges support from the NSF for a Graduate Research Fellowship. M.B.R. acknowledges the NDSEG graduate fellowship program. This work made use of the EPIC facility (NUANCE Center-Northwestern University), which has received support from the MRSEC program (NSF DMR-1121262) at the Materials Research Center; the International Institute for Nanotechnology (IIN); and the State of Illinois, through the IIN.



REFERENCES

(1) Kelly, K. L.; Coronado, E.; Zhao, L. L.; Schatz, G. C. J. Phys. Chem. B 2003, 107, 668−677. (2) Myroshnychenko, V.; Rodriguez-Fernandez, J.; Pastoriza-Santos, I.; Funston, A. M.; Novo, C.; Mulvaney, P.; Liz-Marzan, L. M.; Garcia de Abajo, F. J. Chem. Soc. Rev. 2008, 37, 1792−1805. (3) Scholl, J. A.; Koh, A. L.; Dionne, J. A. Nature 2012, 483, 421− 427. (4) Zhao, Y.; Xu, L.; Ma, W.; Wang, L.; Kuang, H.; Xu, C.; Kotov, N. A. Nano Lett. 2014, 14, 3908−3913. (5) Xia, Y.; Xiong, Y.; Lim, B.; Skrabalak, S. E. Angew. Chem., Int. Ed. 2009, 48, 60−103. (6) Personick, M. L.; Mirkin, C. A. J. Am. Chem. Soc. 2013, 135, 18238−18247. (7) O’Brien, M. N.; Jones, M. R.; Brown, K. A.; Mirkin, C. A. J. Am. Chem. Soc. 2014, 136, 7603−7606. (8) Mayer, K. M.; Hafner, J. H. Chem. Rev. 2011, 111, 3828−3857. (9) Biswas, S.; Liu, X.; Jarrett, J. W.; Brown, D.; Pustovit, V.; Urbas, A.; Knappenberger, K. L.; Nealey, P. F.; Vaia, R. A. Nano Lett. 2015, 15, 1836−1842. (10) Taylor, R. W.; Lee, T.-C.; Scherman, O. A.; Esteban, R.; Aizpurua, J.; Huang, F. M.; Baumberg, J. J.; Mahajan, S. ACS Nano 2011, 5, 3878−3887. (11) Fan, J. A.; Wu, C.; Bao, K.; Bao, J.; Bardhan, R.; Halas, N. J.; Manoharan, V. N.; Nordlander, P.; Shvets, G.; Capasso, F. Science 2010, 328, 1135−1138. (12) Halas, N. J.; Lal, S.; Chang, W.-S.; Link, S.; Nordlander, P. Chem. Rev. 2011, 111, 3913−3961. (13) Jones, M. R.; Osberg, K. D.; MacFarlane, R. J.; Langille, M. R.; Mirkin, C. A. Chem. Rev. 2011, 111, 3736−3827. (14) Moreau, A.; Ciraci, C.; Mock, J. J.; Hill, R. T.; Wang, Q.; Wiley, B. J.; Chilkoti, A.; Smith, D. R. Nature 2012, 492, 86−89. (15) Akselrod, G. M.; Argyropoulos, C.; Hoang, T. B.; Ciraci, C.; Fang, C.; Huang, J. N.; Smith, D. R.; Mikkelsen, M. H. Nat. Photonics 2014, 8, 835−840. (16) Rose, A.; Hoang, T. B.; McGuire, F.; Mock, J. J.; Ciracì, C.; Smith, D. R.; Mikkelsen, M. H. Nano Lett. 2014, 14, 4797−4802. (17) Lee, K.; Asher, S. A. J. Am. Chem. Soc. 2000, 122, 9534−9537. (18) Auguié, B.; Barnes, W. L. Phys. Rev. Lett. 2008, 101, 143902. (19) Kravets, V. G.; Schedin, F.; Grigorenko, A. N. Phys. Rev. Lett. 2008, 101, 087403. (20) Zhou, W.; Odom, T. W. Nat. Nanotechnol. 2011, 6, 423−427. E

DOI: 10.1021/acs.nanolett.5b01548 Nano Lett. XXXX, XXX, XXX−XXX