Nanoplasmonics has recently revolutionized our ability to control light on the nanoscale. Using metallic nanostructures with tailored shapes, it is possible to efficiently focus light into nanoscale field ‘hot spots'. High field enhancement factors have been achieved in such optical nanoantennas, enabling transformative science in the areas of single molecule interactions, highly enhanced nonlinearities and nanoscale waveguiding. Unfortunately, these large enhancements come at the price of high optical losses due to absorption in the metal, severely limiting real-world applications. Via the realization of a novel nanophotonic platform based on dielectric nanostructures to form efficient nanoantennas with ultra-low light-into-heat conversion, here we demonstrate an approach that overcomes these limitations. We show that dimer-like silicon-based single nanoantennas produce both high surface enhanced fluorescence and surface enhanced Raman scattering, while at the same time generating a negligible temperature increase in their hot spots and surrounding environments.

Nanoplasmonics is the emerging research field that studies light-matter interactions mediated by resonant excitations of surface plasmons in metallic nanostructures. It allows the manipulation of the flow of light and its interaction with matter at the nanoscale (10(-9) m). One of the most promising characteristics of plasmonic resonances is that they occur at frequencies corresponding to typical electronic excitations in matter. This leads to the appearance of strong interactions between localized surface plasmons and light emitters (such as molecules, dyes, or quantum dots) placed in the vicinity of metals. Recent advances in nanofabrication and the development of novel concepts in theoretical nanophotonics have opened the way to the design of structures aimed to reduce the lifetime and enhance the decay rate and quantum efficiency of available emitters. In this article, some of the most relevant experimental and theoretical achievements accomplished over the last several years are presented and analyzed.

The absorption process of an emitter close to a plasmonic antenna is enhanced due to strong local electromagnetic (EM) fields. The emission, if resonant with the plasmonic system, re-radiates to the far-field by coupling with the antenna via plasmonic states, whose presence increases the local density of states. Far-field collection of the emission of single molecules close to plasmonic antennas, therefore, provides mixed information of both the local EM field strength and the local density of states. Moreover, super-resolution localizations from these emission-coupled events do not report the real position of the molecules. Here we propose using a fluorescent molecule with a large Stokes shift in order to spectrally decouple the emission from the plasmonic system, leaving the absorption strongly resonant with the antenna's enhanced EM fields. We demonstrate that this technique provides an effective way of mapping the EM field or the local density of states with nanometre spatial resolution.

The formation and subsequent growth of crystalline silicon nanoclusters ͑Si-ncs͒ in annealed silicon-rich silicon oxides ͑SRSOs͒ were studied by glancing angle x-ray diffraction. SRSO samples with Si concentrations ͑y͒ of 0.40, 0.42, and 0.45 were grown by inductively coupled plasma-enhanced chemical-vapor deposition ͑PECVD͒. Samples with y = 0.42 grown by electron-cyclotron-resonance PECVD were also studied. Annealing treatments were performed at temperatures ͑T͒ of 900, 1000, and 1100°C for times ͑t͒ between 0.5 and 3 h in flowing Ar. As-grown SRSO films did not present signs of Si clusters ͑amorphous or crystalline͒; however, ͑111͒, ͑220͒, and ͑311͒ Bragg peaks corresponding to c-Si were clearly seen after annealing at 900°C for the y = 0.45 sample, but only barely seen for the y = 0.42 and undetected for the y = 0.40 samples. For T = 1000°C, all studied SRSO samples clearly showed the c-Si diffraction peaks, which became narrower with increasing t and T. From the width of the Si ͑111͒ peaks, the mean size of Si-ncs and their dependence on T and t was determined. Activation energies were deduced from the T dependence by fitting the results to two growth models of Si precipitates in an a-SiO 2 matrix reported in the literature. The activation energies qualitatively agree with values deduced from transmission electron microscopy studies of annealed SRSO reported in the literature. However, they are significantly lower than Si diffusion activation energies available in the literature for SiO 2 with low excess Si. A broad feature is also observed in the x-ray diffractograms for as-grown samples with low y, which shifts to the peak position corresponding to a-SiO 2 with increasing T. This behavior is explained by the formation of a well-defined a-SiO 2 phase with increasing T, where mixed Si-O 4−n Si n ͑n = 1,2,3͒ tetrahedra in the as-grown alloy are gradually converted into Si-O 4 and Si-Si 4 as phase separation of Si and SiO 2 proceeds. From the measured Si ͑111͒ peak positions, small Si-ncs are found to be tensilely strained by as much as ϳ0.8%. This effect becomes insignificant as Si-ncs become larger with increasing y or T.

Due to the presence of strong magnetic resonances, high refractive index dielectric nanoantennnas have shown the ability to expand the methods available for controlling electromagnetic waves in the subwavelength region. In this work, we experimentally demonstrate that an asymmetric dielectric dimer made of silicon can lead to highly directional scattering depending on the excitation wavelength, due to the interference of the excited magnetic resonances. A back focal plane imaging system combined with a prism coupling technique enables us to explore the scattering profile parallel to the substrate. The directivity of scattering along the substrate is high enough to produce selective guiding of light along the substrate. These results showing tunable control of directional scattering will encourage the realization of novel optical applications, such as optical nanocircuitry.