Experimental Verification of the Spectral Shift between Near- and Far-Field Peak Intensities
of Plasmonic Infrared Nanoantennas
P. Alonso-Gonza´lez,1 P. Albella,1,2 F. Neubrech,3 C. Huck,3 J. Chen (),1,2 F. Golmar,1,4
F. Casanova,1,5 L. E. Hueso,1,5 A. Pucci,3 J. Aizpurua,2 and R. Hillenbrand1,5,*
1CIC nanoGUNE, 20018 Donostia–San Sebastia´n, Spain
2Centro de Fı´sica de Materiales (CSIC-UPV/EHU) and Donostia International Physics Center (DIPC),
20018 Donostia–San Sebastia´n, Spain
3Kirchhoff Institute for Physics, University of Heidelberg, Im Neuenheimer Feld 227, 69120 Heidelberg, Germany
4I.N.T.I.-CONICET, Avenida General Paz 5445, Edificio 42, B1650JKA, San Martı´n, Bs As, Argentina
5IKERBASQUE, Basque Foundation for Science, 48011 Bilbao, Spain
(Received 16 August 2012; published 14 May 2013)
Theory predicts a distinct spectral shift between the near- and far-field optical response of plasmonic
antennas. Here we combine near-field optical microscopy and far-field spectroscopy of individual
infrared-resonant nanoantennas to verify experimentally this spectral shift. Numerical calculations
corroborate our experimental results. We furthermore discuss the implications of this effect in surface-
enhanced infrared spectroscopy.
DOI: 10.1103/PhysRevLett.110.203902 PACS numbers: 42.79.Gn, 42.65.Wi, 73.20.Mf, 78.67.Bf
When a metal nanostructure is illuminated by light, the
excitation of surface plasmons yields strongly concen-
trated optical fields at the metal surface, often referred to
as ‘‘hot spots’’ [1]. Metallic nanostructures can be thus
considered as effective optical nanoantennas for converting
propagating plane waves into localized fields [2]. This
antenna function enables the control of electromagnetic
fields at the nanometer scale [3] and thus has promoted the
development of a vast variety of applications including
surface-enhanced Raman scattering spectroscopy (SERS)
[4–7], surface-enhanced infrared spectroscopy (SEIRS)
[8,9], antenna-enhanced ultrafast nonlinear spectroscopy
[10], near-field microscopy [11–13], and novel photo-
detection schemes [14,15].
To explore new antenna functionalities or to optimize
the antenna performance, different antenna designs have
been developed [16–26]. A major goal is to achieve the
highest local field enhancement, sensitivity, and tunability
of the antennas’ optical response. These properties are
essentially determined by the spectral position and width
of the antenna resonance, which are typically studied by
far-field spectroscopy [27]. However, surface-enhanced
spectroscopies rely on light-matter interaction in the near
field of the antenna, where an object is exposed to high
field intensities in ultrasmall volumes. Early theoretical
studies [28] and recent publications [29–34] predict and
indicate [35] that the spectral near- and far-field response
of the antennas are shifted against each other, which might
have implications for the application and optimization of
optical antennas. Particularly, this spectral shift has been
indicated in SERS [35] and antenna-mediated fluorescence
[36] studies. However, inelastic scattering experiments
present several difficulties such as the inherent difference
between excitation and scattered frequencies or the chemi-
cal bonding and charge transfer between molecules and
metal nanostructures, preventing a rigorous experimental
verification of the shift. Additionally, SERS experiments
usually rely on measurements of samples exhibiting
heterogeneously distributed hot spots of random field
enhancement, which forces a statistical evaluation of the
enhancement. In this work we circumvent these difficulties
by measuring the local near fields of single antennas with
a scattering-type near-field microscope, thus avoiding any
inherent frequency shift between incident and scattered
light or any chemical bonding or charge transfer between
the probe and the antenna. Particularly, we study the
fundamental dipolar mode of antennas resonant in the
midinfrared spectral range and subsequently discuss
implications for surface- and antenna-enhanced infrared
spectroscopy.
The spectral shift between near-field intensity INF
and far-field extinction Iext at the fundamental plasmon
resonance is demonstrated in Fig. 1 for linear dipole
antennas. We show numerical (finite difference in time
domain, Lumerical Solutions) calculations of the near-
and far-field optical response of 40 nm high and 140 nm
wide Au rods of varying length L on a CaF2 substrate.
In Fig. 1(a) we display the spectral positions of the near-
field peak intensity and the far-field extinction peak for
different antenna lengths L. They were obtained by
calculating INF at the antenna extremity [marked by the
red cross in the inset of Fig. 1(a)] and Iext ¼ Iin Itrans,
respectively, as a function of the illumination wavelength
. Iin is the incident intensity and Itrans is the calculated
transmitted intensity in the far field of the antennas.
The polarization of the incident light was parallel to
the antenna axes. A spectral shift between the near-field
and far-field peak intensities is observed throughout the
whole spectral range from visible to midinfrared wave-
lengths. At a fixed antenna length L, the near-field peak is
PRL 110, 203902 (2013) P HY S I CA L R EV I EW LE T T E R S week ending17 MAY 2013
0031-9007=13=110(20)=203902(6) 203902-1 2013 American Physical Society

shifted to a longer wavelength compared to the far-field
extinction maximum. This shift is shown in more detail
in Fig. 1(b), where we plot Iext and INF along the vertical
dashed line in Fig. 1(a) (antenna length L ¼ 3:1 m).
Considering a fixed illumination wavelength , the near-
field peak appears at smaller L compared to the far-field
extinction. We illustrate this effect in Fig. 1(c) by plot-
ting Iext and INF along the horizontal line in Fig. 1(a)
( ¼ 9:3 m).
Based on Mie theory for small spheres, the shift between
near- and far-field peak intensities was already studied by
Messinger et al. in the 1980s [28]; however, only recently
this phenomenon has been intuitively explained by describ-
ing metallic antennas as classically driven harmonic oscil-
lators [32,33]. When the harmonic oscillator is damped
(which can be associated with dissipation in the antenna
and with scattering losses), the maximum oscillation
amplitude (which can be associated with the near-field
amplitude) appears at lower frequencies than the maximum
dissipation (which can be associated with the far-field
absorption). While the spectral position of the maximum
dissipation does not depend on the damping, the maximum
oscillation amplitude generally shifts to lower energies
with increasing damping. This explains why spectral shifts
between near- and far-field peak intensities can be quite
large in the case of strong plasmon damping, for example
in plasmonic Ni antennas [33] or in strongly scattering
antennas [37].
Here we experimentally verify the shift between near-
and far-field peak intensities for midinfrared antennas.
To that end, we measure at a fixed wavelength both the
near-field intensity [Fig. 2(a)] and far-field extinction
[Fig. 2(b)] of individual antennas of varying length L,
thus tracing the resonances according to Fig. 1(c). This
approach is used because quantitative near-field data are
readily obtained by recording a single near-field image
[Fig. 3(a)] of an antenna set where L is systematically
varied [38–40]. Furthermore, in applications such as in
SEIRS, the antenna length is the essential parameter to
(a)
(b)
Si tip ItransIin
INF
Iin
FIG. 2 (color online). Illustration of the detection schemes
employed for measuring the near- and far-field response of
individual infrared antennas: (a) s-SNOM, and (b) infrared
microspectroscopy.
(a)
(b) (c)
Iext
x INF
FIG. 1 (color online). Numerical study of the near-field inten-
sity INF and far-field extinction Iext of linear dipole Au antennas
on a CaF2 substrate. (a) Spectral positions of the near-field peak
intensity and the far-field extinction peak. The lines are guides
to the eye. The inset illustrates where the near-field intensity and
the far-field extinction were evaluated. (b) Near-field intensity
INF and far-field extinction Iext as a function of wavelength for
an antenna length L ¼ 3:1 m [spectra along the vertical dashed
line in (a)]. (c) Near-field intensity INF and far-field extinction
Iext as a function of nanorod length L for a fixed illumination
wavelength ¼ 9:3 m [spectra along the horizontal solid line
in (a)]. The spectra in (b) and (c) were normalized to their
maximum values.
(a)
0
M
ax
5µm
A B
EDC
(b)
FIG. 3 (color online). Experimental near- and far-field study of
linear infrared dipole antennas. (a) s-SNOM image of the nano-
antennas at ¼ 9:3 m. From the top to the bottom and from
the left to the right, the antenna length increases from L ¼ 2 m
to L ¼ 4:4 m. The cross indicates the position where the
near-field intensities displayed in Fig. 4 have been measured.
(b) Far-field extinction spectra of the antennas marked in
Fig. 3(a) by the letters A-E. The vertical dashed line indicates
the wavelength ¼ 9:3 m, where the extinction values dis-
played in Fig. 4 have been extracted.
PRL 110, 203902 (2013) P HY S I CA L R EV I EW LE T T E R S week ending17 MAY 2013
203902-2

be matched to the fixed vibrational resonance of the mole-
cules under study [9,41].
The antennas are Au rods of 40 nm height, 140 nm
width, and a length varying from L ¼ 2 m to L ¼
4:4 m, fabricated by electron beam lithography on a
CaF2 substrate. The distance between the different anten-
nas is 10 m, which allows for measuring a far-field
extinction spectrum of each individual antenna, as well
as for recording a single near-field image of the whole
antenna set [Fig. 3(a)]. We thus obtain for each individual
antenna both near-field intensity and far-field extinction,
allowing for a quantitative comparison of the two quanti-
ties as a function of the antenna length L.
Near-field imaging [Fig. 2(a)] is performed with a side-
illumination scattering-type scanning near-field optical
microscope (s-SNOM, from Neaspec GmbH). A Si tip
oscillating vertically at frequency is used for locally
scattering the antenna near fields [21,33,38,39,42–44].
Both tip and antenna are illuminated with s-polarized
infrared light from a CO2 laser at an angle of 50 from
the surface normal. The light scattered by the tip is
recorded with a pseudoheterodyne interferometer [45].
By locating a polarizer in front of the detector, we select
the horizontally polarized scattered light. Demodulation
of the detector signal at a higher-harmonic frequency n
yields background free near-field signals [21,43]. We note
that by (i) illuminating the antennas with s-polarized light
and (ii) detecting the s-polarized backscattered light, the
demodulated amplitude signal sn yields the square of the
local near-field amplitude [39], sn ¼ INF.
Far-field spectroscopy [Fig. 2(b)] is performed with an
infrared microscope (Bruker Hyperion 1000) coupled to a
Fourier transform spectrometer (Bruker Tensor 27), yield-
ing infrared extinction spectra of the same set of individual
antennas. The antennas are illuminated with thermal radia-
tion (polarization parallel to the long axis of the antenna) of
intensity Iin under normal incidence. To address individual
antennas, the illumination is through a 10 10 m size
aperture. The transmitted light Itrans is recorded with a reso-
lution of 8 cm1, yielding far-field extinction spectra
Iext ¼ Iin Itrans. These are normalized to the extinction
measured at least 30 m away from the antennas (refer-
ence spectrum).
Figure 3(a) shows the near-field image of the antenna set,
revealing the typical dipolar mode pattern for each antenna
(two bright spots, indicating the strongly concentrated near
fields at the rod extremities) [42]. The near-field signal
increases with increasing antenna length, until it reaches
its maximum at L3:1m (marked by position C). With
further increasing antenna length, the near-field signal
decreases. This observation clearly reveals the resonance
behavior of the antennas [46]. In Fig. 3(b) we show the far-
field extinction spectra of the individual antennas marked in
Fig. 3(a).We observe how the far-field resonance (extinction
maximum) shifts to longer wavelengths when the antenna
length increases, following the typical behavior of dipole
antennas [9].
To compare the near-field and far-field optical responses
of the antennas, we plot in Fig. 4 both INF and Iext as a
function of the antenna length L for the fixed illumination
wavelength ¼ 9:3 m. The near-field intensities have
been extracted from Fig. 3(a) at the extremities of the
antennas where the maximum value is obtained (the cross
in Fig. 3(a) marks the typical position). We point out that
we study the fundamental antenna mode, which implies
that the near-field spectrum is the same at every point on
the surface of the antenna. The extinction has been
extracted from the individual far-field extinction spectra
of the antennas, some of them shown in Fig. 3(b). We
clearly see in Fig. 4 that the near-field and far-field peak
intensities are shifted against each other. The far-field
extinction maximum occurs at L ¼ 3:30 m (marked by
the black dashed line in Fig. 4), while the near-field maxi-
mum appears at a shorter antenna length of L ¼ 3:10 m
(marked by the red dashed line in Fig. 4). These experi-
mental results indeed verify the shift between the calcu-
lated near-field intensity spectrum INF (red line in Fig. 4,
obtained at the antennas’ extremities as in Fig. 1) and the
calculated far-field extinction spectrum Iext ¼ Iin Itrans
(black curve in Fig. 4, obtained in the far field of the
antennas as in Fig. 1) of the antennas. We note that no
fitting of the numerical results has been applied, just
normalization of all the curves to their maximum values.
We also note that the use of dielectric tips together with the
s-polarized sample illumination has been shown to faith-
fully measure the spectral near-field response of plasmonic
antennas without introducing spectral shifts [47–49]. We
thus can exclude that the significant spectral shift between
the near- and far-field response of the antennas is intro-
duced by the tip.
FIG. 4 (color online). Experimental and calculated near-field
intensity INF (red) and far-field extinction Iext (black) as a
function of the antenna length. The experimental near-field
intensities were measured at the extremity of the individual
nanoantennas, at the position indicated by a cross in Fig. 3(a).
The experimental far-field extinction values were extracted from
the far-field spectra at ¼ 9:3 m, as indicated by a vertical
dashed line in Fig. 3(b). The red solid line INF shows the
numerically calculated near-field intensity at the rod extremity
[as indicated by the inset in Fig. 1(a)]. The black solid line Iext
shows the numerically calculated total extinction evaluated in
the far field of the antennas. Both experimental and calculated
data were normalized to the corresponding maximum values.
PRL 110, 203902 (2013) P HY S I CA L R EV I EW LE T T E R S week ending17 MAY 2013
203902-3

The shift between near- and far-field peak intensities may
have important implications for sensing applications, as the
optical interaction between molecules and antennas is
mediated by the near field. The spectroscopic information
about the molecules, however, is measured in the far field.
In order to elucidate the influence of the shift in antenna-
enhanced infrared extinction spectroscopy, we performed
a numerical study of the near- and far-field response of
molecules in the vicinity of infrared antennas. We consider
Au antennas covered with a 20-nm-thick layer of poly-
methyl methacrylate (PMMA) molecules on a CaF2
substrate. PMMA is used as an example because of its
well-defined infrared vibrational resonance. Figure 5(a)
shows the calculated extinction IPMMAext of PMMA-coated
antennas of different lengthsL (black curves).With increas-
ing L, the antenna resonance shifts to longer wavelengths,
whereas the vibrational response of PMMA appears fixed at
¼ 5:8 m, independently of L. Importantly, the spectral
PMMA response is enhanced when it is close to the antenna
resonance. Simultaneously, its line shape is modified, which
results from the Fano-like interference of both infrared
resonances [9,41,50–52]. To isolate and quantify the
antenna-enhanced spectral response of PMMA, we show
in Fig. 5(b) the difference spectraC ¼ IPMMAext Iext, where
Iext [green curves in Fig. 5(a)] is the resonance of the
antenna when covered by a 20-nm-thick reference layer
with a dielectric value "¼2:1 (baseline subtraction). We
now define the ‘‘fingerprint contrast’’ C as the difference
between the maximum and the minimum values of each
individual spectrum C [illustrated by the schematics in
Fig. 5(b)]. In Fig. 5(c) we showC for the different antenna
lengths L. We find that the largest fingerprint contrast is
obtained for L ¼ 1:7 m. By comparing C with the
near-field intensity and far-field extinction at the vibrational
resonance of PMMA [both shown in Fig. 5(d)], we interest-
ingly observe that the largest fingerprint contrast is obtained
when the near-field intensity IPMMANF (rather than the far-field
extinction IPMMAext ) reaches its maximum. This can be under-
stood when considering (i) that the molecular layer absorbs
energy from the antenna’s near field rather from its far
field and (ii) that the absorption of the molecules is much
stronger than their scattering. The spectral shift between
near- and far-field peak intensities thus has to be considered
when optimizing antennas for spectroscopy applications.
In Fig. 5(e) we compare the near-field spectrum (INF, red
dashed line) and the far-field spectrum (IPMMAext , black solid
line) of the 1.7-mm-long PMMA-coated antennas, where
the fingerprint contrast in the far-field extinction spectrum
is largest. We observe that the near-field intensity peak of
the antenna matches the molecular vibration, rather than
the far-field extinction maximum. To see this more clearly,
we show the near-field intensity INF (red solid line) and
far-field extinction Iext (green solid line). In addition to the
spectral shift, we observe that the spectral shape of the
molecular fingerprint also differs in the near-field and far-
field signals. We note that further extended experimental
(a) (b) (c)
(d)
(e)
I
ext
PMMA
I
ext
*
Au
CaF2L
CaF2
PMMA
*
Au
FIG. 5 (color online). Numerical study of antenna-enhanced extinction spectroscopy of polymethyl methacrylate molecules.
(a) Left: Illustration of the spectroscopy configuration and dielectric functions of the PMMA and reference layers, "PMMA and ".
Right: Extinction spectra of PMMA-coated Au antennas on a CaF2 substrate for different lengths L (IPMMAext , black curves). The green
curves Iext show the extinction spectra of the antennas covered by a 20-nm-thick reference layer with a dielectric value ".
(b) Difference spectra C ¼ IPMMAext Iext. (c) ‘‘Fingerprint contrast’’ C as defined in the scheme of (b). (d) Near-field intensity
(IPMMANF , red curve) and far-field extinction (IPMMAext , black curve) of PMMA-coated Au antennas as a function of antenna length L. The
illumination wavelength is ¼ 5:8 m, matching the vibrational resonance of the PMMA molecules. (e) Near-field intensity (IPMMANF ,
red dashed curve) and far-field extinction (IPMMAext , black solid curve) of PMMA-coated Au antennas as a function of illumination
wavelength . For comparison, we also show the near-field intensity INF (red solid line) and far-field extinction Iext (green solid line).
The antenna length is L ¼ 1:7 m, corresponding to the maximum of C in (c).
PRL 110, 203902 (2013) P HY S I CA L R EV I EW LE T T E R S week ending17 MAY 2013
203902-4

and theoretical studies are needed for more detailed
insights into this phenomenon, as well as to draw general
conclusions regarding the optimal spectral contrast, since a
variety of antenna designs, molecule vibrations, and physi-
cal spectroscopy processes (extinction, Raman scattering,
fluorescence) exists. Nevertheless, the canonical case of a
dipolar antenna for SEIRS, as studied in this work, already
stresses the importance and necessity of considering spec-
tral shifts between near- and far-field peaks, in order to
optimize surface- and antenna-enhanced spectroscopies.
In conclusion, having performed experimental studies of
individual infrared-resonant antennas by near-field micro-
scopy and far-field extinction spectroscopy, we confirm
experimentally the spectral shift between the near- and
far-field peak intensities. Furthermore, we have studied
numerically the implications of this spectral shift in
SEIRS, showing that it has to be considered in order to
optimize the molecular spectral absorption contrast in
plasmonic (bio)sensing devices.
This work was supported by the European FP7 Project
‘‘Nanoantenna’’ (No. FP7-HEALTH-F5-2009-241818-
NANOANTENNA), the Spanish National Projects
No. MAT2009-08398 and No. FIS2010-19609-C02-01,
and the Basque Government Projects ETORTEK
(nanoIKER) and excellence program IT756-13.
*r.hillenbrand@nanogune.eu
[1] H. Xu, J. Aizpurua, M. Ka¨ll, and P. Apell, Phys. Rev. E 62,
4318 (2000).
[2] L. Novotny and N. van Hulst, Nat. Photonics 5, 83 (2011).
[3] J. A. Schuller, E. S. Barnard, W. Cai, Y. C. Jun, J. S. White,
and M. L. Brongersma, Nat. Mater. 9, 193 (2010).
[4] D. L. Jeanmaire and R. P. Van Duyne, J. Electroanal.
Chem. Interfacial Electrochem. 84, 1 (1977).
[5] M. Moskovits, Rev. Mod. Phys. 57, 783 (1985).
[6] K. Kneipp, Y. Wang, H. Kneipp, L.T. Perelman, I. Itzkan,
R.R. Dasari, andM.S. Feld, Phys. Rev. Lett. 78, 1667 (1997).
[7] S. Nie and S. R. Emory, Science 275, 1102 (1997).
[8] M. Osawa, Bull. Chem. Soc. Jpn. 70, 2861 (1997).
[9] F. Neubrech, A. Pucci, T.W. Cornelius, S. Karim, A.
Garcı´a-Etxarri, and J. Aizpurua, Phys. Rev. Lett. 101,
157403 (2008).
[10] T. Schumacher, K. Kratzer, D. Molnar, M. Hentschel,
H. Giessen, and M. Lippitz, Nat. Commun. 2, 333 (2011).
[11] P. Anger, P. Bharadwaj, and L. Novotny, Phys. Rev. Lett.
96, 113002 (2006).
[12] S. Ku¨hn, U. Ha˚kanson, L. Rogobete, and V. Sandoghdar,
Phys. Rev. Lett. 97, 017402 (2006).
[13] R. Hillenbrand and F. Keilmann, Appl. Phys. B 73, 239
(2001).
[14] L. Tang, S. E. Kocabas, S. Latif, A.K. Okyay, D.-S.
Ly-Gagnon, K. C. Saraswat, and D.A. B. Miller, Nat.
Photonics 2, 226 (2008).
[15] M.W. Knight, H. Sobhani, P. Nordlander, and N. J. Halas,
Science 332, 702 (2011).
[16] C. Radloff and N. J. Halas, Nano Lett. 4, 1323 (2004).
[17] A. G. Curto, G. Volpe, T. H. Taminiau, M. P. Kreuzer,
R. Quidant, and N. F. van Hulst, Science 329, 930 (2010).
[18] C.L. Nehl, H. Liao, and J.H. Hafner, Nano Lett. 6, 683 (2006).
[19] J. Pe´rez-Juste, I. Pastoriza-Santos, L.M. Liz-Marza´n, and
P. Mulvaney, Coord. Chem. Rev. 249, 1870 (2005).
[20] J. A. Fan, C. Wu, K. Bao, J. Bao, R. Bardhan, N. J. Halas,
V. N. Manoharan, P. Nordlander, G. Shvets, and F.
Capasso, Science 328, 1135 (2010).
[21] P. Alonso-Gonzalez, M. Schnell, P. Sarriugarte, H.
Sobhani, C. Wu, N. Arju, A. Khanikaev, F. Golmar,
P. Albella, L. Arzubiaga, F. Casanova, L. E. Hueso, P.
Nordlander, G. Shvets, and R. Hillenbrand, Nano Lett.
11, 3922 (2011).
[22] L. Chuntonov and G. Haran, Nano Lett. 11, 2440 (2011).
[23] N. Verellen, P. Van Dorpe, C. Huang, K. Lodewijks,
G. A. E. Vandenbosch, L. Lagae, and V.V. Moshchalkov,
Nano Lett. 11, 391 (2011).
[24] S. Mukherjee, H. Sobhani, J. B. Lassiter, R. Bardhan, P.
Nordlander, and N. J. Halas, Nano Lett. 10, 2694 (2010).
[25] M. Hentschel, M. Saliba, R. Vogelgesang, H. Giessen,
A. P. Alivisatos, and N. Liu, Nano Lett. 10, 2721 (2010).
[26] M. Pelton, J. Aizpurua, and G. Bryant, Laser Photonics
Rev. 2, 136 (2008).
[27] P. Zijlstra, P.M. R. Paulo, and M. Orrit, Nat. Nanotechnol.
7, 379 (2012).
[28] B. J. Messinger, K.U. von Raben, R.K. Chang, and
P.W. Barber, Phys. Rev. B 24, 649 (1981).
[29] S. Bruzzone, M. Malvaldi, G. P. Arrighini, and C.
Guidotti, J. Phys. Chem. B 110, 11050 (2006).
[30] N. K. Grady, N. J. Halas, and P. Nordlander, Chem. Phys.
Lett. 399, 167 (2004).
[31] B.M. Ross and L. P. Lee, Opt. Lett. 34, 896 (2009).
[32] J. Zuloaga and P. Nordlander, Nano Lett. 11, 1280 (2011).
[33] J. Chen, P. Albella, Z. Pirzadeh, P. Alonso-Gonza´lez,
F. Huth, S. Bonetti, V. Bonanni, J. A˚ kerman, J. Nogue´s,
P. Vavassori, A. Dmitriev, J. Aizpurua, and R. Hillenbrand,
Small 7, 2341 (2011).
[34] M. Kats, N. Yu, P. Genevet, Z. Gaburro, and F. Capasso,
Opt. Express 19, 21748 (2011).
[35] A. D. McFarland, M.A. Young, J. A. Dieringer, and
R. P. Van Duyne, J. Phys. Chem. B 109, 11279 (2005).
[36] V. Giannini, G. Vecchi, and J. Go´mez Rivaz, Phys. Rev.
Lett. 105, 266801 (2010).
[37] G.W. Bryant, F. J. Garcia de Abajo, and J. Aizpurua, Nano
Lett. 8, 631 (2008).
[38] J. Dorfmu¨ller, R. Vogelgesang, R. T. Weitz, C. Rockstuhl,
C. Etrich, T. Pertsch, F. Lederer, and K. Kern, Nano Lett.
9, 2372 (2009).
[39] P. Alonso-Gonza´lez, P. Albella, M. Schnell, J. Chen,
F. Huth, A. Garcı´a-Etxarri, F. Casanova, F. Golmar, L.
Arzubiaga, L. E. Hueso, J. Aizpurua, and R. Hillenbrand,
Nat. Commun. 3, 684 (2012).
[40] P. Alonso-Gonza´lez, P. Albella, F. Golmar, L. Arzubiaga,
F. Casanova, L. Hueso, J. Aizpurua, and R. Hillenbrand,
Opt. Express 21, 1270 (2013).
[41] V. Giannini, Y. Francescato, H. Amrania, C. C. Phillips,
and S.A. Maier, Nano Lett. 11, 2835 (2011).
[42] M. Schnell, A. Garcia Etxarri, A. J. Huber, K. B. Crozier,
J. Aizpurua, and R. Hillenbrand, Nat. Photonics 3, 287
(2009).
[43] M. Schnell, A. Garcia-Etxarri, J. Alkorta, J. Aizpurua, and
R. Hillenbrand, Nano Lett. 10, 3524 (2010).
PRL 110, 203902 (2013) P HY S I CA L R EV I EW LE T T E R S week ending17 MAY 2013
203902-5

[44] J. Dorfmu¨ller, D. Dregely, M. Esslinger, W. Khunsin,
R. Vogelgesang, K. Kern, and H. Giessen, Nano Lett.
11, 2819 (2011).
[45] N. Ocelic, A. Huber, and R. Hillenbrand, Appl. Phys. Lett.
89, 101124 (2006).
[46] P. Biagioni, M. Savoini, J.-S. Huang, L. Duo`, M. Finazzi,
and B. Hecht, Phys. Rev. B 80, 153409 (2009).
[47] A. Garcı´a-Etxarri, I. Romero, F. Javier Garcı´a de Abajo,
R. Hillenbrand, and J. Aizpurua, Phys. Rev. B 79, 125439
(2009).
[48] J. Dorfmu¨ller, R. Ogelgesang, R. Thomas Weitz, C.
Rockstuhl, C. Etrich, T. Pertsch, F. Lederer, and K.
Kern, Nano Lett. 9, 2372 (2009).
[49] M. Schnell, A. Garcia-Etxarri, A. J. Huber, K. B. Crozier,
A. Borisov, J. Aizpurua, and R. Hillenbrand, J. Phys.
Chem. C 114, 7341 (2010).
[50] B. Gallinet and O. J. F. Martin, ACS Nano 5, 8999 (2011).
[51] B. Gallinet and O. J. F. Martin, Opt. Express 19, 22167 (2011).
[52] J. Aizpurua, T. Taubner, F. J. Garcı´a de Abajo, M. Brehm,
and R. Hillenbrand, Opt. Express 16, 1529 (2008).
PRL 110, 203902 (2013) P HY S I CA L R EV I EW LE T T E R S week ending17 MAY 2013
203902-6