The practical efficacy of technologically promising metals for use in ultraviolet plasmonics (3−6 eV) is assessed by an exhaustive numerical analysis. This begins with estimates of the near-and far-field electromagnetic enhancement factors of isolated hemispherical and spherical metallic nanoparticles deposited on typical dielectric substrates like sapphire, from which the potential of each metal for plasmonic applications may be ascertained. The ultraviolet plasmonic behavior of aluminum, chromium, copper, gallium, indium, magnesium, palladium, platinum, rhodium, ruthenium, titanium, and tungsten was compared with the well-known behavior of gold and silver in the visible. After exploring this behavior for each metal as a function of nanoparticle shape and size, the deleterious effect caused by the metal's native oxide is considered, and the potential for applications such as surface-enhanced Raman spectroscopy, accelerated photodegradation and photocatalysis is addressed.

The nonoxidizing catalytic noble metal rhodium is introduced for ultraviolet plasmonics. Planar tripods of 8 nm Rh nanoparticles, synthesized by a modified polyol reduction method, have a calculated local surface plasmon resonance near 330 nm. By attaching p-aminothiophenol, local field-enhanced Raman spectra and accelerated photodamage were observed under near-resonant ultraviolet illumination, while charge transfer simultaneously increased fluorescence for up to 13 min. The combined local field enhancement and charge transfer demonstrate essential steps toward plasmonically enhanced ultraviolet photocatalysis.

Abstract:The ultraviolet (UV) range presents new challenges for plasmonics, with interesting applications ranging from engineering to biology. In previous research, gallium, aluminum, and magnesium were found to be very promising UV plasmonic metals. However, a native oxide shell surrounds nanostructures of these metals that affects their plasmonic response. Here, through a nanoparticle-oxide core-shell model, we present a detailed electromagnetic analysis of how oxidation alters the UV-plasmonic response of spherical or hemisphere-on-substrate nanostructures made of those metals by analyzing the spectral evolution of two parameters: the absorption efficiency (far-field analysis) and the enhancement of the local intensity averaged over the nanoparticle surface (near-field analysis).

Self-assembled, irregular ensembles of hemispherical Ga nanoparticles (NPs) were deposited on sapphire by molecular beam epitaxy. These samples, whose constituent unimodal or bimodal distribution of NP sizes was controlled by deposition time, exhibited localized surface plasmon resonances tunable from the ultraviolet to the visible (UV/ vis) spectral range. The optical response of each sample was characterized using a variable-angle spectroscopic ellipsometer, and the dielectric response of the ensemble of NPs on each sample was parametrized using Lorentz oscillators. From this, a relationship was found between NP size and the deduced Lorentzian parameters (resonant frequency, damping, oscillator strength) for most unimodal and bimodal samples at most frequencies and angles of incidence. However, for samples with a bimodal size distribution, Mueller matrix ellipsometry revealed nonspecular scattering at particular frequencies and angles, suggesting a resonant interparticle coupling effect consistent with recently observed strong local field enhancements in the ultraviolet. KEYWORDS: plasmonics, ultraviolet, gallium, nanoparticles, variable-angle spectroscopic ellipsometry, variable-angle Mueller matrix ellipsometry R esearch into the optical response of confined electronic plasma oscillations in metallic nanostructures has become the foundation of modern plasmonics. Metallic nanostructures are routinely fabricated with desired spatial geometries for specific applications including single-molecule detection, enhanced fluorescence, toxic remediation, and catalysis. 1−6 Because the optical response depends sensitively on nanostructure composition, size, and shape, increasingly sophisticated characterization tools are required: aggregate behavior may be characterized by absorption spectroscopy or dark-field microscopy, while single-particle behavior may be characterized by confocal microscopy, near-field scanning optical microscopy, and cathodoluminescence. 7−12 These tools provide incredibly detailed understanding of the absorption and scattering properties of plasmonic metal nanoparticles, especially when complemented by sophisticated electromagnetic modeling techniques such as finite difference time domain, finite element codes, and the discrete dipole approximation. 13−19 Of growing interest is the need to monitor the plasmonic properties of substrate-supported nanoparticle (NP) ensembles during fabrication. Because it is impractical to use single-particle imaging tools for this application, advances in the more traditional tools are needed to understand single-particle behaviors from aggregate measurements. Spectroscopic ellipsometry (SE) has been used to monitor how NP ensembles alter the polarization state of an incident light beam as a function of wavelength and incidence angle. 20−24 Recently, a fixed incidence angle spectroscopic ellipsometer mounted on a molecular beam epitaxy (MBE) chamber has been used to monitor in real time the deposition of nanometer-scale metallic films and NPs through...

The influence of the degree of purity of a silicon nanoparticle on its resonances, either electric or magnetic, is assessed by using Mie theory as well as finite-element simulations. In particular, it is shown that the main effect of the increase of absorption due to the pollutants is observed in the magnetic resonances. Concerning Kerker's conditions for the directionality of the scattering [J. Opt. Soc. Am.73, 765 (1983)], it is found that both are strongly shifted when the material's purity is varied. Resistive losses confirm the quenching of magnetic resonances, showing that the region of influence in the magnetic dipole resonance is much larger than in the electric one, although it has been found that losses are not critical for silicon content over 99.50%.