18944 Phys. Chem. Chem. Phys., 2013, 15, 18944--18950 This journal is c the Owner Societies 2013

Cite this: Phys. Chem.Chem.Phys.,2013,15, 18944

Near-field spatial mapping of strongly interactingmultiple plasmonic infrared antennas†

Sarah E. Grefe,a Daan Leiva,a Stefan Mastel,b Scott D. Dhuey,c Stefano Cabrini,c

P. James Schuckc and Yohannes Abate*a

Near-field dipolar plasmon interactions of multiple infrared antenna structures in the strong coupling

limit are studied using scattering-type scanning near-field optical microscope (s-SNOM) and theoretical

finite-difference time-domain (FDTD) calculations. We monitor in real-space the evolution of plasmon

dipolar mode of a stationary antenna structure as multiple resonantly matched dipolar plasmon

particles are closely approaching it. Interparticle separation, length and polarization dependent studies

show that the cross geometry structure favors strong interparticle charge–charge, dipole–dipole and

charge–dipole Coulomb interactions in the nanometer scale gap region, which results in strong field

enhancement in cross-bowties and further allows these structures to be used as polarization filters.

The nanoscale local field amplitude and phase maps show that due to strong interparticle Coulomb

coupling, cross-bowtie structures redistribute and highly enhance the out-of-plane (perpendicular to the

plane of the sample) plasmon near-field component at the gap region relative to ordinary bowties.

1 Introduction

Plasmonic optical antennas lead to strong localization andenhancement of electromagnetic radiation below the diffractionlimit.1–5 They are at the heart of some of the most innovativetechnological applications based on manipulation of photophysicalprocesses at the nanometer scale. These include high-resolutionmicroscopy and spectroscopy, solar energy conversion, photo-catalysis, nanoscale optical circuits and quantum optics andcommunication.6–11

Since the local plasmon near-field distribution determinesthe effectiveness of antennas, their investigation has been thefocus of several studies using techniques that involve eitherelectrons12,13 or photons.14–20 These techniques have exploreda number of parameters, including imaging of geometry-relatedspatial patterns of antenna resonances, intensity characterizationof hotspots,4,21,22 energy-resolved plasmon eigenmodes,23,24 multi-photon photoluminescence,17,19 harmonic generation or frequencymixing,18,25,26 and emission patterns of nanoantennas.27,28

To achieve improved structures with superior function,designs that involve multiple nanoantenna components thatare near-field coupled to each other are required.22,24,29–31

Various theoretical models have been proposed to elucidatenear-field plasmon interactions.32–34 A fascinating analogy hasbeen made between plasmon coupling and atomic orbitalhybridization well-known in molecular physics.32,35 Beginningwith the pioneering work of Prodan et al.32 the plasmonhybridization model has been used to explain plasmon resonancesin metal nanoshells,15,32,36,37 thin metallic films,38,39 nano-stars,40 spherical dimers,41 rod dimers,22,24,30 and nanoparticleaggregates,42 mainly in the optical frequency regime. Individualparticle plasmon resonance modes corresponding to discreteorbital angular momentum values then form hybridized modeswith other particles or cavities, depending on the geometry.However, direct experimental studies of the near-field interactionsin the nanometer gap regime are difficult due to limitationsimposed by both nanofabrication and nanocharacterizationtechniques. In particular, the nanoscale interactions of severalplasmonic structures and the spatial evolution of the fielddistributions in real space remain mostly unexplored.4,20,43–47

Here we perform polarization-selective scattering-type scanningnear-field optical microscope (s-SNOM) measurements to directlyvisualize the dipolar near-field distributions and couplings ofplasmonic triangles, bowties and cross-bowtie antennas inthe mid-infrared (9–11 mm) spectral regime. We systematicallyinvestigate the coupling mechanism by first imaging the near-fielddistribution of a single triangular antenna. While monitoring its

a Department of Physics and Astronomy, California State University Long Beach,

Long Beach, California 90840, USA. E-mail: Yohannes.Abate@csulb.edub CIC NanoGUNE Consolider, Tolosa Hiribidea 76, 20018 Donostia San Sebastian,

Spainc The Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley,

California 94720, USA

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c3cp53104j

Received 23rd July 2013,Accepted 9th August 2013

DOI: 10.1039/c3cp53104j

www.rsc.org/pccp

PCCP

PAPER

Publ

ishe

d on

13

Sept

embe

r 20

13. D

ownl

oade

d by

Geo

rgia

Sta

te U

nive

rsity

on

31/0

1/20

17 2

0:16

:22.

View Article OnlineView Journal | View Issue

This journal is c the Owner Societies 2013 Phys. Chem. Chem. Phys., 2013, 15, 18944--18950 18945

field distribution, we add another identical antenna closer andcloser to it (essentially forming bowties of varying gap width). Wethen cross the existing bowtie with another bowtie (forming a cross-bowtie at varying proximity distances) and monitor the evolution ofthe strongly coupled dipolar field distribution modified throughinter-antenna charge–dipole and dipole–dipole Coulomb inter-actions by performing polarization control measurements.

The triangular plasmonic metal nanostructures concentrateplasmons into the sharp end and smear them over the widenedbase, resulting in dipolar modes with higher field intensity atthe sharp end, and a largely suppressed field at the base.48–51

We find that the aspect ratio, the base and sharp end widthdetermine the symmetry and intensity of localized field profiles,both for single particles and particle aggregates.48,50,51

Employing the cross-polarized S/P s-SNOM imaging scheme(exciting along the in-plane direction, S-excitation, and detectingthe out-of-plane near-field component, P-detection), we furtherdemonstrate that cross-bowtie structures could be used as activeinfrared polarization filters. Such devices would be of interestto realize antenna structures for surface-enhanced infraredabsorption spectroscopy, to manipulate optical fields in subwave-length field distributions, plasmonic filtering and polarizationselection applications.

2 Experimental

The experimental setup is a commercial s-SNOM system (Nea-SNOM, neaspec.com), which has been described before.20,52,53

Briefly, a probing CO2 laser is linearly S polarized throughoutthe experiment and is focused on the tip–sample interface at anangle of B501 from the sample surface. The scattered field isacquired using a pseudoheterodyne interferometric detectionscheme. Suppression of the background signal is achieved byvertical tip oscillation at the cantilever’s mechanical resonancefrequency (O B 285 kHz) and demodulation of the detectorsignal at higher harmonics nO. In this work, second harmonicdemodulation (n = 2) is used for all experiments. The combinationof the scattered field from the tip and the reference beam passesthrough a linear polarizer which further selects the P polarizationof the measured signal for analysis (see ESI†).

The Au plasmonic samples are fabricated via electron-beamlithography as a grid of B26 nm thick single triangle, bowtie orcross-bowtie structures atop a 3 nm Ti adhesion layer on aSiO2 substrate. Each structure in the grid has systematicallyand symmetrically incremented the interparticle distance(B57–137 nm) and length (B1284–1766 nm), covering a widerange of possible geometric combinations.

Experimental results are theoretically supported by finite-difference time-domain (FDTD) simulations (Lumerical Inc.,lumerical.com). For all simulations (unless otherwise specified),each Au triangle was modeled by a trapezoidal polygon, possessinga 26 nm thickness, 50 nm sharp tip width, 450 nm wide trianglebase, and 1800 nm length. These dimensions were averagedfrom topography scan measurements. Each particle is simulatedatop a SiO2 substrate. The optical excitation source is a mid-infrared (9–11 mm) plane wave.

3 Results and discussions

Triangular antenna structures offer an extra degree of controlsince the hotspot field strength and spatial localization at theends of the antenna can be controlled by varying the widths ofthe sharp and broad ends. Fig. 1 shows experimental topography,near-field amplitude and phase images of a single triangularantenna. The optical near-field amplitude and phase images wereacquired using the S/P scheme, with excitation wavelength fixedat l = 10.5 mm, and the polarization direction fixed along thelongitudinal axis of the antenna structure. The amplitude imagedisplays strong optical contrast at the sharp end and at the base,while the phase image shows 1801 phase shift between each halfof the triangle, confirming a dipolar resonance mode.45,54

We performed FDTD calculations to explore the structuraldependence of the plasmon field distribution across the antenna.Fig. 2(a) shows FDTD near-field intensity (|Ez|

2) images, calculatedfrom the out-of-plane (perpendicular to the sample plane)

Fig. 1 Experimental topography (coordinate axes shown at top left), near-fieldamplitude (s2), (excitation direction along white arrow) and phase (f2) images ofa single triangular antenna, with excitation wavelength fixed at l = 10.5 mm. Thelength of the antenna is 1284 nm.

Fig. 2 (a) FDTD intensity (|Ez|2) calculations at l = 10.5 mm of a single Au particleevolving from a rod to triangular structures, the polarization direction along thewhite arrow. (b) Corresponding field intensity line profiles of the images shown in(a). (c) Plot of intensity at the position marked ‘‘X’’ on the antenna structures as afunction of base width. The inset (1) represents a rod structure at 50 nm widthand (2) represents experimentally used base width (450 nm), coordinate axesdefined at bottom right.

Paper PCCP

Publ

ishe

d on

13

Sept

embe

r 20

13. D

ownl

oade

d by

Geo

rgia

Sta

te U

nive

rsity

on

31/0

1/20

17 2

0:16

:22.

View Article Online

18946 Phys. Chem. Chem. Phys., 2013, 15, 18944--18950 This journal is c the Owner Societies 2013

Ez field, and corresponding line profiles extracted across thelongitudinal axis (Fig. 2(b)). The width of the sharp end and thelength of the rod are fixed at 50 nm and 1800 nm respectively.

As the width of the rod at one end increases, the intensitymeasured at the fixed end, marked in Fig. 2(a), decreasesdramatically, while intensity at the expanding end decreaseswith a weak resonance peak at around 75 nm. When the basewidth of the triangle is equal to the fixed end width, the particleis essentially a rod, with a symmetric optical amplitude profileas shown in the inset in Fig. 2(c).17,29,30,48,50 Increasing thewidth of the base shifts the resonance peak to lower energy andlocalizes the strong near-fields at the ends of the triangleasymmetrically (see ESI†). Spectroscopic FDTD calculationsshow the higher amplitude signal at the tip than at the baseof the triangles, which is a combination of strong field concen-tration at the tip and shift of resonance frequency as the basewidens (see ESI†). These results indicate the importance ofstructure optimization for resonant hotspot localization in suchplasmonic structures.

Fig. 3 displays experimental topography, optical near-fieldamplitude and phase images of a bowtie (Fig. 3(a)) and cross-bowtie (Fig. 3(b)) antenna structures. The amplitude and phaseimages were acquired at excitation wavelength fixed at l =10.5 mm, and the incident laser polarization direction is fixedalong the longitudinal bowtie axis of the structures as shown bythe white arrow in the amplitude image of Fig. 3(a). Opticalimages display no contrast in the gap region as a result ofP-polarization selective detection. Expected dipolar patterns areobserved along the polarization axis where large near-fieldamplitude and 1801 phase shift at the rod extremities whereasthe bowtie oriented transverse to the polarization axis showsmuch weaker optical contrast. These contrast formations anddependence on interparticle separation and antenna length areexplored in detail below.

In Fig. 4 we first systematically investigate using FDTD theeffect of mutual interactions on the near-field intensity (|Ez|

2)of a single antenna structure as bowtie and cross-bowtie

structures are brought in close proximity to it. Line plots inFig. 4(a)–(d) show FDTD calculations of the localized intensityas a function of interparticle separations calculated at thelocation ‘‘X’’ in each corresponding colored illustration (bottomof Fig. 4(a)–(d)). All single antenna structures have identicalgeometric dimensions (length 1800 nm, base width 450 nm,sharp tip width 50 nm) and the excitation polarization is orientedhorizontally as shown in Fig. 4(d) by the double sided blackarrow. The interparticle separation is measured along the blackarrow shown in illustration, Fig. 4(a), (b) and (d). As a baseline forcomparison, we show the constant intensity of a single triangularantenna structure at position ‘‘X’’ in Fig. 4(a), which can bethought of as an infinite sized gap bowtie (Fig. 4, dotted blackline). The red dash-dotted curve (Fig. 4(b)) shows the intensity ofa stationary triangle at the position ‘‘X’’ as a function of theseparation with a second triangle. The comparison of curve(a) and curve (b) shows how the field intensity on a stationarystructure is modified due to an approaching antenna. In thecross-bowtie structures, the minimum gap width is limited bythe sharp tip widths. As a baseline for comparison, we show theconstant maximum intensity value taken at a 60 nm gap bowtieshown by the green straight dashed line in Fig. 4(c), which maybe considered to be a cross-bowtie antenna with infinite separa-tion between vertical arm pairs. The solid blue curve in Fig. 4(d)shows the stationary bowtie near-field intensity modificationtaken at position ‘‘X’’ as a function of the vertical arm pairinterparticle separation. These calculations show strong inter-particle separation dependence on the field intensity and thatcross antennas redistribute the fields, putting more field inten-sity near their gaps compared to bowties. Animation showing theevolution of the near-field intensity of a cross-bowtie as multipleantenna structures are assembled is shown in the ESI.†

Fig. 3 Mid-infrared (l = 10.5 mm) s-SNOM spatial-mapping of infrared antennasshowing topography (left, coordinate axes shown at bottom right), second-harmonic amplitude (middle, also shown excitation direction along white arrow)and the second-harmonic phase (right). The length of each triangular antennas in(a) 1690 nm and in (b) 1766 nm.

Fig. 4 Plots (a)–(d) show FDTD calculations of the localized intensity (|Ez|2)calculated at the red ‘‘X’’ position in each corresponding illustration (a)–(d). Allarms have identical geometric dimensions (length 1800 nm, base width 450 nm,sharp tip width 50 nm) as shown on the coordinate axes and the black doublesided arrow.

PCCP Paper

Publ

ishe

d on

13

Sept

embe

r 20

13. D

ownl

oade

d by

Geo

rgia

Sta

te U

nive

rsity

on

31/0

1/20

17 2

0:16

:22.

View Article Online

This journal is c the Owner Societies 2013 Phys. Chem. Chem. Phys., 2013, 15, 18944--18950 18947

Experimentally, we further explore this proximity effect onthe plasmon near-field distribution by imaging several similargeometry cross-bowties with varying gap width. Fig. 5 comparesexperimental s-SNOM topography, second harmonic amplitude(s2) and phase (f2) images of the smallest and largest gap widthcross-bowties imaged, 57 nm gap width (Fig. 5(a)) and 137 nmgap width cross-bowties (Fig. 5(b)). In both amplitude andphase images, the bowtie along the polarization direction isresonantly excited maintaining dipolar modes in each arm forthe gap widths studied here.54 The cross-bowtie arm transverseto the polarization axis displays weak contrast with asymmetricsplitting along the arm length. Without a priori knowledge tothe polarization direction, the optical images could clearlyindicate the incident beam polarization direction, showingthat the cross antenna structures could serve as polarizationindicators. The asymmetric splitting in the optical images is aresult of slight misalignment of the sample with the excitationpolarization; with a perfect alignment the splitting should besymmetric about the bowtie axis (see ESI† for FDTD calculationimages). Further simulation results showing the evolution ofthe near-field intensity of a cross-bowtie as the sample isrotated with respect to a fixed polarization of the light sourceis shown in the ESI.†

The amplitude images clearly show stronger optical contrastfor a 57 nm gap width cross-bowtie compared to that of a

137 nm gap cross-bowtie. In Fig. 5(c), quantitative experimentaloptical amplitude signals (normalized to the SiO2 substrate)taken at the position marked by a green ‘‘X’’ in Fig. 5(a) and (b)are plotted as a function of measured gap widths, demonstrat-ing the increase in amplitude contrast as the spacing betweenthe cluster of particles decreases. This result is fully reproducedby FDTD calculations of the field intensity as a function of gapwidth shown in Fig. 5(d), using similar dimension cross-bowtiesto that for the experiment at excitation wavelength, l = 10.5 mm.Calculations for bowties are also shown in the same figure forcomparison, revealing for the same gap width stronger intensityin cross-bowties. We note that in Fig. 5(d) the cross-bowtiesmallest gap width is limited by the width of the sharp tip.

We systematically further explore and quantify the outstandingcapability of cross-bowties to enhance light in the gap regioncompared with single triangular antennas or bowties by performingcontrolled experiments and FDTD calculations. In Fig. 6(a)–(c)we show experimental normalized (to SiO2 substrate) near-fieldamplitude signal average (normalized to their minima) taken atthe sharp end of a triangular antenna when it is isolated orforming a bowtie and a cross-bowtie antenna structures as afunction of antenna length. In Fig. 6(e) we show amplitude mapsof the antenna array for these measurements, while in Fig. 6(d)

Fig. 5 (a) and (b) Experimental topography (left, coordinate axes shown at left),near-field optical amplitude s2 (middle, white arrow shows excitation direction),and phase f2 (right) with gap sizes labeled on the topography images in nm.(c) Experimental s2 values measured at the green ‘‘X’’ position indicated in (a)and (b); inset showing schematics of how the gap sizes in the experiments andcalculations are measured. (d) FDTD calculations of near-field intensity (|Ez|2) atthe same location ‘‘X’’ as gap size increases.

Fig. 6 Experimental near-field amplitude (s2) signal values measured at thesharp tip of the antennas excited using S-polarized laser at wavelength l =10.5 mm for (a) single triangular antennas, (b) bowties, (c) cross-bowties, (d) FDTDsimulation field intensity (|Ez|2) at similar parameters as the experiment.(e) Experimental amplitude (s2) images on which results shown (a)–(c) are based.The experimental amplitude data are normalized to the SiO2 substrate andcorresponding minimum values. The scale bar in (e) represents 1 mm.

Paper PCCP

Publ

ishe

d on

13

Sept

embe

r 20

13. D

ownl

oade

d by

Geo

rgia

Sta

te U

nive

rsity

on

31/0

1/20

17 2

0:16

:22.

View Article Online

18948 Phys. Chem. Chem. Phys., 2013, 15, 18944--18950 This journal is c the Owner Societies 2013

we show FDTD simulation of the field intensity using the sameparameters as that for the experiments. Since the length rangestudied in our experiment is smaller than the characteristicresonance length at l = 10.5 mm, antenna field enhancementand so the normalized near-field amplitude contrast increaseswith length for all the structures shown in Fig. 6(a) and (b) whichis also recently reported by Brown et al.55 We observe that thefield intensity and so the dipolar amplitude contrast and itsincrease with length (comparing slopes in Fig. 6) are largest incross-bowtie structures compared to bowties or single antennastructures as shown experimentally (Fig. 6(a)) and by FDTDcalculations (Fig. 6(b)).

Both the gap dependence (Fig. 4 and 5) and the lengthdependence (Fig. 6) studies indicate that the field enhancementin cross-bowtie antenna structures is modified strongly byinter-antenna mutual interactions. To understand the natureof the interaction around the gap region better we performedFDTD simulation of the charge density distributions of theantenna structures with similar geometry and gap width to theexperiments and compare their charge densities with eachother and with their corresponding field intensities. Fig. 7shows a zoom-in map of the charge densities of a singletriangular antenna (Fig. 7(a)), a bowtie (Fig. 7(b)), and a cross-bowtie (Fig. 7(c)), and the corresponding field intensity imagesare shown in Fig. 7(d)–(f). Charge density distribution iscalculated by taking the difference of Ez 2 nm above and2 nm below the top surface of structures, which is proportionalto charge density by Gauss’ Law. The spatial charge densityimages show strong dipolar charge localization along thepolarization axis (shown in Fig. 6(a) with positive chargesmarked blue and negative charges red). The bowtie transverseto the polarization axis in Fig. 7(c) also shows charge distribu-tion along the width of the bowtie with a dipole oriented alongthe excitation polarization direction. Comparison of the chargedensity values of the three structures at the locations marked by‘‘X’’ in Fig. 7(a)–(c) reveals that the charge density on a cross-bowtie is 1.2 times than that of a bowtie and 2.5 times than thatof an isolated triangular antenna. This corresponds to anintensity on the cross-bowtie which is 1.4 times than that of abowtie and 3.9 times than that of a an isolated triangularantenna. The extra charge on the cross-bowtie at the antennatip location shown in Fig. 7(c) is due to dipole–dipole, charge–dipole and charge–charge complex interactions at dielectric

proximity on a nanometer scale with the transverse bowtie. Inaddition to the effect of the dipole–dipole interaction of theother longitudinal arm of the bowtie, the charges at the tips ofthe bowtie oriented along the polarization axis feel attractivecharge–charge, charge–dipole and dipole–dipole Coulomb inter-actions from both the upper and lower arms of the transversebowtie. This is clearly shown in the charge density calculation inFig. 7(c), where charge–charge interactions occur between theblue positively charged regions on the left side of the verticalarms and neighboring red negatively charged left bowtie arm,and similarly for the right side regions of opposite charges.Along the horizontal axis, each vertical arm forms a dipole,which also interacts with the charges on the individual neigh-boring bowtie arms. Moreover, the dipole formed across thehorizontal bowtie gap interacts with the charges on the verticalbowtie arms. These additional Coulomb interactions pumpmore charge to the tips of the antennas generating strong lightconfinement at the gap region. As the gap width decreases, themutual dipolar Coulomb coupling increases resulting in largeramplitude contrast as shown in Fig. 5. Inter-antenna Coulombinteraction becomes weak and so is the near-field enhancementas the triangular antennas are pulled apart from each other. Asthe length of the triangular antennas increase, Coulomb inter-actions between closely confined antenna structures alsoincrease which in turn increases the field intensity associatedwith each antenna. Much larger enhancement could be found inthe immediate area surrounding the tips of the antenna rods byforming such antenna structures using longer rods or tuning theexcitation laser to their resonance peaks.

4 Conclusion

S-SNOM imaging and FDTD simulations show that triangularplasmonic antenna structures produce asymmetrical dipolarplasmon resonances, characterized by a highly confined fieldintensity hotspot at the sharp end compared to the wider base.Using cross-polarized s-SNOM amplitude contrast, we quantifyplasmon nanoscale interactions of multiple infrared antennastructures. We find that the detected P component of theamplitude in cross-bowties shows a stronger signal comparedto bowties with similar gap width and geometry as evidence ofmultiple dipole–dipole and charge–dipole Coulomb interactions.The spatial evolution of the near-field intensity and polarizationdependence in cross-bowties reveal that these structures act asinfrared polarization selectors. Future wide wavelength rangespectroscopic imaging experiments using broadband lightsource and varying particle geometry dependent studies, as wellas the in-plane detection scheme where the hotspot in the gapregion can be mapped in the coupling region of the antennastructures, will help to further elucidate the full details of theobserved plasmonic coupling multiple interactions we present.

Acknowledgements

This work was supported by the U.S. Army Research Office,Agreement Number: W911NF-12-1-0076. S. E. G. is grateful to

Fig. 7 FDTD simulations of (a)–(c) the real part of charge density distribution,(d)–(f) corresponding intensity images. In (a) the black arrow shows the excitationdirection.

PCCP Paper

Publ

ishe

d on

13

Sept

embe

r 20

13. D

ownl

oade

d by

Geo

rgia

Sta

te U

nive

rsity

on

31/0

1/20

17 2

0:16

:22.

View Article Online

This journal is c the Owner Societies 2013 Phys. Chem. Chem. Phys., 2013, 15, 18944--18950 18949

the California State University Long Beach Graduate ResearchFellowship for support. Work at the Molecular Foundry, LawrenceBerkeley National Laboratory was supported by the Director,Office of Science, Office of Basic Energy Sciences, Division ofMaterials Sciences and Engineering, of the U.S. Department ofEnergy under Contract No. DE-AC02-05CH11231.

References

1 P. J. Schuck, D. P. Fromm, A. Sundaramurthy, G. S. Kino andW. E. Moerner, Phys. Rev. Lett., 2005, 94, 017402.

2 S. Eustis and M. A. El-Sayed, Chem. Soc. Rev., 2006, 35, 209–217.3 P. Bharadwaj, B. Deutsch and L. Novotny, Adv. Opt. Photonics,

2009, 1, 438–483.4 S. Mastel, S. E. Grefe, G. B. Cross, A. Taber, S. Dhuey,

S. Cabrini, P. J. Schuck and Y. Abate, Appl. Phys. Lett.,2012, 101, 131102.

5 R. D. Grober, R. J. Schoelkopf and D. E. Prober, Appl. Phys.Lett., 1997, 70, 1354–1356.

6 T. H. Taminiau, J. R. R. J. Moerland, F. B. Segerink,L. Kuipers and N. F. van Hulst, Nano Lett., 2007, 7, 28–33.

7 M. Thomas, J.-J. Greffet, R. Carminati and J. R. Arias-Gonzalez, Appl. Phys. Lett., 2004, 85, 3863–3865.

8 K. Nakayama, K. Tanabe and H. A. Atwater, Appl. Phys. Lett.,2008, 93, 121904.

9 S. Pillai, K. R. Catchpole, T. Trupke, G. Zhang, J. Zhao andM. A. Green, Appl. Phys. Lett., 2006, 88, 161102.

10 A. Kinkhabwala, Z. Yu, S. Fan, Y. Avlasevich, K. Mullen andW. E. Moerner, Nat. Photonics, 2009, 3, 654–657.

11 A. Alu and N. Engheta, Nat. Photonics, 2008, 2, 307–310.12 A. L. Koh, K. Bao, I. Khan, W. E. Smith, G. Kothleitner,

P. Nordlander, S. A. Maier and D. W. McComb, ACS Nano,2009, 3, 3015–3022.

13 I. Alber, W. Sigle, F. Demming-Janssen, R. Neumann,C. Trautmann, P. A. van Aken and M. E. Toimil-Molares,ACS Nano, 2012, 6, 9711–9717.

14 L. Novotny and N. van Hulst, Nat. Photonics, 2011, 5, 83–90.15 H. Wang, Y. Wu, B. Lassiter, C. L. Nehl, J. H. Hafner,

P. Nordlander and N. J. Halas, Proc. Natl. Acad. Sci.U. S. A., 2006, 103, 10856–10860.

16 N. Verellen, Y. Sonnefraud, H. Sobhani, F. Hao,V. V. Moshchalkov, P. V. Dorpe, P. Nordlander andS. A. Maier, Nano Lett., 2009, 9, 1663–1667.

17 D.-S. Wang, F.-Y. Hsu and C.-W. Lin, Opt. Express, 2009, 17,11350–11359.

18 K. D. Ko, A. Kumar, K. H. Fung, R. Ambekar, G. L. Liu,N. X. Fang and K. C. Toussaint, Nano Lett., 2011, 11, 61–65.

19 P. Biagioni, D. Brida, J.-S. Huang, J. Kern, L. Duo, B. Hecht,M. Finazzi and G. Cerullo, Nano Lett., 2012, 12, 2941–2947.

20 J. M. Stiegler, Y. Abate, A. Cvitkovic, Y. E. Romanyuk,A. J. Huber, S. R. Leone and R. Hillenbrand, ACS Nano,2011, 5, 6494–6499.

21 P. Alonso-Gonzalez, P. Albella, M. Schnell, J. Chen, F. Huth,A. Garcıa-Etxarri, F. Casanova, F. Golmar, L. Arzubiaga,L. Hueso, J. Aizpurua and R. Hillenbrand, Nat. Commun.,2012, 3, 1–7.

22 P. Alonso-Gonzalez, P. Albella, F. Golmar, L. Arzubiaga,F. Casanova, L. E. Hueso, J. Aizpurua and R. Hillenbrand,Opt. Express, 2013, 21, 1270–1280.

23 B. Schaffer, U. Hohenester, A. Trugler and F. Hofer, Phys.Rev. B: Condens. Matter Mater. Phys., 2009, 79, 041401.

24 J.-S. Huang, J. Kern, P. Geisler, P. Weinmann, M. Kamp,A. Forchel, P. Biagioni and B. Hecht, Nano Lett., 2010, 10,2105–2110.

25 U. K. Chettiar and N. Engheta, Phys. Rev. B: Condens. MatterMater. Phys., 2012, 86, 075405.

26 S. Palomba and L. Novotny, Nano Lett., 2009, 9, 3801–3804.27 T. Kosako, Y. KAdoya and H. F. Hofmann, Nat. Photonics,

2010, 4, 312–315.28 I. S. Maksymov, A. R. Davoyan and Y. S. Kivshar, Appl. Phys.

Lett., 2011, 99, 083304.29 D. Weber, P. Albella, P. Alonso-Gonzalez, F. Neubrech,

H. Gui, T. Nagao, R. Hillenbrand, J. Aizpurua andA. Pucci, Opt. Express, 2011, 19, 15047–15061.

30 A. M. Funston, C. Novo, T. J. Davis and P. Mulvaney, NanoLett., 2009, 9, 1651–1658.

31 D. P. Fromm, A. Sundaramurthy, P. J. Schuck, G. Kino andW. E. Moerner, Nano Lett., 2004, 4, 957–961.

32 E. Prodan, C. Radloff, N. J. Halas and P. Nordlander, Science,2003, 302, 419–422.

33 L. Novotny, Phys. Rev. Lett., 2007, 98, 266802.34 D. Schurig, J. B. Pendry and D. R. Smith, Opt. Express, 2006,

14, 9794–9804.35 N. J. Halas, S. Lal, W. S. Chang, S. Link and P. Nordlander,

Chem. Rev., 2011, 111, 3913–3961.36 N. K. Grady, N. J. Halas and P. Nordlander, Chem. Phys. Lett.,

2004, 399, 167–171.37 E. Prodan and P. Nordlander, J. Chem. Phys., 2004, 120,

5444–5454.38 A. Aubry, D. Y. Lei, S. A. Maier and J. B. Pendry, ACS Nano,

2011, 5, 3293–3308.39 Y. Fang and Y. Huang, Appl. Phys. Lett., 2013, 102, 153108.40 F. Hao, C. L. Nehl, J. H. Hafner and P. Nordlander, Nano

Lett., 2007, 7, 729–732.41 P. Nordlander, C. Oubre, E. Prodan, K. Li and

M. I. Stockman, Nano Lett., 2004, 4, 899–903.42 N. A. Mirin, K. Bao and P. Nordlander, J. Phys. Chem. A,

2009, 113, 4028–4034.43 H.-Y. Lin, C.-H. Huang, C.-H. Chang, Y.-C. Lan and

H.-C. Chui, Opt. Express, 2010, 18, 165–172.44 M. Schnell, A. Garcıa-Etxarri, A. J. Huber, K. Crozier, J. Aizpurua

and R. Hillenbrand, Nat. Photonics, 2009, 3, 287–291.45 D.-S. Kim and Z. H. Kim, Opt. Express, 2012, 20, 8689–8699.46 M. Zavelani-Rossi, M. Celebrano, P. Biagioni, D. Polli,

M. Finazzi, L. Duo, G. Cerullo, M. Labardi, M. Allegrini,J. Grand and P.-M. Adam, Appl. Phys. Lett., 2008, 92, 093119.

47 Z. Nuno, B. Hessler, J. Ochoa, Y.-S. Shon, C. Bonney andY. Abate, Opt. Express, 2011, 19, 20865–20875.

48 H. Fischer and O. J. F. Martin, Opt. Express, 2008, 16, 9144–9154.49 M. Schnell, P. Alonso-Gonzalez, L. Arzubiaga, F. Casanova,

L. E. Hueso, A. Chuvilin and R. Hillenbrand, Nat. Photonics,2011, 5, 283–287.

Paper PCCP

Publ

ishe

d on

13

Sept

embe

r 20

13. D

ownl

oade

d by

Geo

rgia

Sta

te U

nive

rsity

on

31/0

1/20

17 2

0:16

:22.

View Article Online

18950 Phys. Chem. Chem. Phys., 2013, 15, 18944--18950 This journal is c the Owner Societies 2013

50 N. Yu, E. Cubukcu, L. Diehl, D. Bour, S. Corzine, J. Zhu,G. Hofler, K. B. Crozier and F. Capasso, Opt. Express, 2007,15, 13272.

51 Y.-Y. Yang, Y.-L. Zhang, F. Jin, X.-Z. Dong, X.-M. Duan andZ.-S. Zhao, Opt. Commun., 2011, 284, 3474–3478.

52 F. Keilmann and R. Hillenbrand, Phil. Trans. R. Soc., A, 2004,362, 787–805.

53 R. Hillenbrand and F. Keilmann, Phys. Rev. Lett., 2000, 85,3029–3032.

54 D.-S. Kim, J. Heo, S.-H. Ahn, S. W. Han, W. S. Yun andZ. H. Kim, Nano Lett., 2009, 9, 3619–3625.

55 L. V. Brown, K. Zhao, N. King, H. Sobhani, P. Nordlanderand N. J. Halas, J. Am. Chem. Soc., 2013, 135,3688–3695.

PCCP Paper

Publ

ishe

d on

13

Sept

embe

r 20

13. D

ownl

oade

d by

Geo

rgia

Sta

te U

nive

rsity

on

31/0

1/20

17 2

0:16

:22.

View Article Online