news & views
NATURE NANOTECHNOLOGY | VOL 13 | JUNE 2018 | 437–443 | www.nature.com/naturenanotechnology
© 2018 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
if needed, the alignment process due to
the current lateral error of 10 μ m in the
automated alignment) and how to ensure
an environmentally controlled atmosphere
during the whole process (in the current
implementation some of the processes — the
exfoliation of the flakes and the transfer of
the assembled stack onto the final acceptor
substrate — are carried out in air, which
represents a serious drawback when working
with air-sensitive materials). Another long-
term challenge is the operating complexity
of the transfer system for potential users.
This issue remains to be resolved despite
the effort on the authors’ part to provide
a comprehensive description of the set-up
and even to share the software in an open
repository. Indeed, the recent experience
demonstrated that the success of many
experimental techniques, such as the
mechanical exfoliation of 2D materials,
strongly relies on their ability to be easily
adopted by other groups. Yet despite the
remaining issues, the demonstrated robotic
set-up represents an important step towards
the realization of 2D-based devices with
arbitrary complexity. ❐
Instituto Madrileño de Estudios Avanzados en
Nanociencia (IMDEA Nanociencia), Campus de
Cantoblanco, Madrid, Spain.
Factory, Instituto de Ciencia de Materiales de Madrid
(ICMM-CSIC), Madrid, Spain.
*e-mail: email@example.com; andres.
Published online: 14 May 2018
1. Will robots displace humans as motorised vehicles ousted horses?
e Economist (1 April 2017).
2. Acemoglu, D. & Restrepo, P. e Race Between Machine and
Man: Implications of Technology for Growth, Factor Shares and
Employment (National Bureau of Economic Research, 2016).
3. Masubuchi, S. et al. Nat. Commun. 9, 1413 (2018).
4. Geim, A. K. & Grigorieva, I. V. Nature 499,
5. Dean, C. R. et al. Nat. Nanotech. 5, 722–726 (2010).
6. Frisenda, R. et al. Chem. Soc. Rev. 47, 53–68 (2018).
7. Withers, F. et al. Nat. Mater. 14, 301–306 (2015).
Conﬁning light to the atomic scale
A graphene sheet near a metal nanoantenna squeezes infrared photons into a subnanometric gap, pushing the
limits of nanophotonics.
Alberto G. Curto and Jaime Gómez Rivas
he future of light-based technologies
relies on the miniaturization of optical
components to achieve faster, more
efficient and more sensitive optoelectronic
devices. It hinges on our ability to shrink
light in at least one dimension while
allowing it to propagate in other directions.
One approach to confine photons to a
scale much smaller than the free-space
) consists of forming surface
plasmon-polaritons (electromagnetic waves
bound to a metal–dielectric interface).
However, there is a trade-off: strong spatial
confinement results in shorter propagation
lengths. The origin of the relation between
confinement and losses is the Landau
damping, the excitation by the tightly
confined surface wave of electron–hole
pairs in the metal that causes loss of energy
stored in the plasmon. Now, writing in
Science, Alcaraz Iranzo et al. have managed
to confine light to the ultimate limit, that is,
an atomically thin layer, while guiding it for
hundreds of nanometres
Despite the inherent limitations imposed
by the Landau damping, a metal–insulator–
metal nanocavity (MIM) was used as
early as 2006 to compress visible light
into a 3-nm-gap
, but at the expense of it
propagating for only a few oscillations. For
applications requiring confinement but
where propagation losses are not an issue,
antenna-on-a-mirror platforms (a variant
of the MIM geometry also known as patch
antennas) can provide confinement to a
nanometre-scale gap in localized hotspots
Graphene can also sustain surface
plasmon waves, now in the mid-infrared
at wavelengths around 10 μ m. Through a
gate voltage, the standing wave resonances
associated with these propagating plasmons
can be electrically tuned by changing the
Fermi energy of graphene. Plasmonic
resonators that exploit this effect have been
made out of graphene nanoribbons, but
the electric field is not confined to the one-
nanometre scale and oscillations barely exist
over the extent of the plasmon wavelength
Following an intermediate route to
confinement, Alcaraz Iranzo et al. use an
antenna-on-graphene geometry. More
specifically, the sample consists of a
graphene layer separated by a thin dielectric
spacer from metal nanoantennas with cavity
sizes between 33 and 256 nm (Fig. 1).
In this way, they can harness at the same
time the large cross-section of metals
for efficient excitation and the superior
confinement of graphene plasmons.
The advantage of two-dimensional (2D)
materials over bulk metals is that they screen
electric fields differently. They exhibit a
non-local response with a strong
momentum dependence of the dielectric
constant. The presence of a metal near
the graphene layer screens the graphene
plasmons, and the gap size provides out-of-
In general, for a surface plasmon, the
propagation length (L
) quantifies the
decay in the propagation direction and it is
related to the imaginary part of the plasmon
wavevector (k), L
= 1/|2Im(k)|, whereas the
plasmon wavelength λ
= 2π /Re(k) reflects
the oscillation period. In a surface plasmon
cavity, the resonances are at approximate
cavity lengths w = mλ
/2, where m is an
integer. An appropriate figure of merit
for a resonator is thus the normalized
propagation length L
which gives the number of oscillations
that occur before attenuation. Therefore,
the number of observed resonances in a
spectrum is a proxy for the confinement–
In the experiments of Alcaraz Iranzo
et al., the appearance of higher-order Fabry–
Pérot modes is a signature of confined but
propagating light, to a degree not seen in
previous studies of graphene plasmons.
Using far-field excitation and detection in
a Fourier-transform infrared spectrometer,
the authors acquire extinction spectra.
By turning on and off the screened graphene
plasmons with a voltage and analysing the
differential transmittance, they can show up