news & views NOBEL PRIZE IN CHEMISTRY
Seeing the nanoscale 2 μm
© 2000 NAS, USA
Microscopy and nanotechnology, convention tells us, should not mix. The prefixes on the words themselves give some indication of their intrinsic disparity — a factor of one thousand separates the micro- and the nanoworlds. The 2014 Nobel Prize in Chemistry has been awarded to Stefan Hell, William Moerner and Eric Belzig for helping to bridge this gap with their development of super-resolved fluorescence microscopy. Light is perhaps our most important connection with the world around us. Seeing is believing, or so they say, and direct visual observation remains our most trusted means of scientific discovery and experimental verification. But light has its limits. Diffraction means that a lens cannot focus a ray to a spot smaller than between one third and one half of the wavelength. This is just a few hundred nanometres for a conventional visible-light microscope: good enough to image whole cells, but not to get a detailed view of what is going on inside them. In 1994, Stefan Hell began developing a concept that he thought could beat this limit called stimulated emission depletion (STED) microscopy. Hell’s idea was to take advantage of fluorescence quenching — a reduction in the intensity of light emitted from a molecule when it is exposed to too much laser light. One laser stimulates fluorescence while
a second quenches all emission except for that at the centre of the focus spot. In 2000, Hell experimentally demonstrated a STED microscope with a focus spot six times smaller than that imposed by the diffraction limit. He and his team were able to image an Escherichia coli bacterium at a resolution not possible in a conventional microscope (pictured; the image on the left was obtained with standard confocal resolution and the image on the right with the axial resolution improved by STED; T. A. Klar et al., Proc. Natl Acad. Sci. USA 97, 8206–8210; 2000). Around a similar time, William Moerner was trying to isolate the emission from single molecules. He dispersed the green fluorescent protein from a jellyfish in a gel and was able to collect light from just one protein when the protein–protein separation was greater than the diffraction limit. In 1997, Moerner used this approach to discover that the emission from these
proteins blinks on and off. In 2006, Eric Belzig harnessed this effect for imaging. He and his co-workers combined molecules that emitted light of different colours and superimposed images taken over time while the emitters switched on and off. They used this single-molecule microscope to image intracellular proteins with a nanometre resolution. STED and single-molecule microscopy are already reaping rewards. The three Nobel laureates themselves have used their techniques to better understand brain synapses, cell division in embryos and the proteins associated with Huntington’s disease. But the real testament to the importance of their work is the extent to which the methods have now been adopted and improved by their colleagues around the world, and the promise of fundamental discoveries yet to come. DAVID GEVAUX
OPTOMECHANICS
Photons that pivot and shuttle
Self-oscillation and shuttling of photons between distinct cavities in a photonic see-saw provide unexpected opportunities in nano-optomechanics.
Heedeuk Shin and Peter T. Rakich
I
n the past few years, micro- and nanoscale systems have enabled controllable coupling between photons and acoustic phonons in a variety of forms1,2. Quantum3,4 and classical nonlinear interactions5, produced by coupling between tightly confined optical and mechanical modes, have spawned a flurry of activity in the field 878
of optomechanics. Recent experiments have demonstrated the versatility of such device physics; optomechanical interactions have been used to implement schemes for cooling of phonon modes, optomechanical memory, signal delay and self-oscillation1,2,6. The rapidly growing field of optomechanics has produced a host of
device topologies that have enabled strong photon–phonon coupling. These include optomechanical systems based on suspended mirrors, membranes, microtoroids, nanobeams, microspheres, microdisks, waveguides and photonic crystals2. To control such interactions with ever greater fidelity, both ultralow-loss phonon modes
NATURE NANOTECHNOLOGY | VOL 9 | NOVEMBER 2014 | www.nature.com/naturenanotechnology
© 2014 Macmillan Publishers Limited. All rights reserved
news & views and greatly enhanced optical forces provide a boost in performance; these key attributes simultaneously reduce the impedance of a vibrational mode while increasing the strength of the photon–phonon coupling. As a consequence, phonons can be kicked into oscillation with an ever decreasing number of photons to produce long-lived phononic excitations. A new frontier of optomechanics will likely include more complex forms of classical and quantum information processing. To realize such systems, methods for coupling numerous optomechanical systems will be necessary to permit extended forms of coherence7–9. Chip-scale nanooptomechanical systems are a promising platform for such concepts, as a multitude of nano-optomechanical systems can be controllably coupled on a single chip. Now, writing in Nature Nanotechnology, Huan Li and Mo Li from the University of Minnesota demonstrate a new form of intracavity optomechanical coupling using extended excitations in a purposely designed cavity optomechanical system10. It consists of a nanobeam within which two photonic crystal nanocavities are embedded. These optical cavities are symmetrically placed across the central pivot point of the nanobeam, so that the beam can only oscillate vertically. When one cavity moves closer to the substrate, the other moves further away, yielding a torsional movement analogous to a see-saw.
Air
c
y
e θ>0
Si Si
Si SiO2
SiO2 Ex field
νLaser
Intensity
Trans.
b
are produced, the computed mode-fields within the cross-section of a nanobeam are seen in Fig. 1a alongside a sketch of the corresponding intensity profile (Fig. 1b). Because it is energetically favourable for the electromagnetic energy to reside in materials of higher dielectric constant, photons in a cavity act to pull the dielectric substrate closer to the cavity mode (Fig. 1c). In other words, radiation pressure acts to pull the dielectric substrate closer to the tightly bound cavity mode, much the same way that optical tweezers pull dielectric particles in the focus of a Gaussian beam (Fig. 1d). In this see-saw system, the displacement of each cavity is anti-correlated. Through interaction of each cavity mode with the substrate, the dynamical tuning of the cavity modes are 180° out of phase as the torsional mode oscillates. As a consequence of the anti-correlated cavity detuning, a new form of photon shuttling 17 ensues: modulation of intercavity coupling, produced by the torsional motion, permits dynamical photon tunnelling between the two spatially distinct cavities. Through experiments, laser light of fixed frequency (vertical green line in Fig. 1e–g) is evanescently coupled to the left cavity producing self-oscillation. During self-oscillation, the basic concept of the photon shuttling can be understood as follows: for torsional states θ > 0 and θ = 0,
SiO2 Optical force
Frequency
f
θ=0
Trans.
a
The authors show that optomechanical (or light-driven) self-oscillation is produced as light is injected into either nanocavity. This self-oscillation drives the torsional modes of the dual cavity system into action. The strong interaction between the photonic modes in each nanocavity and the substrate produces an anti-correlated tuning of the resonant frequency in each cavity, resulting in unique dynamics. Moreover, the low dissipation of the torsional mode and strong optomechanical coupling produced by these cavities enable self-oscillation at exceedingly low (0.14 μW) threshold powers. The observed self-oscillation results from strong optical forces produced by photons within the cavity system. In siliconbased nano-optomechanics, two types of optical force can provide mechanical motion, electrostrictive forces and radiation pressure11,12. Electrostrictive forces are produced by the dynamical response of media to light 13. Radiation-pressure-induced forces are entirely geometric in nature, and result from the change in momentum of photons as they reflect or scatter from material boundaries14. Strong photon–phonon coupling in this see-saw system is mediated almost exclusively by radiation pressure; changes in local geometry — produced by the motion of the cavity relative to the substrate — act to shift the energy of nanocavity modes15,16. To illustrate how these forces
d
Frequency
θ 0 (e), θ = 0 (f) and θ