The lycurgus cup

Probably the first historic report of plasmonic phenomena and arguably one of the most beautiful is the Lycurgus cup (British Museum, London, UK). Under ambient lighting, this fourth-century Roman chalice appears green, but when illuminated from the inside, the transmitted light shimmers bright red. Today we know that this fascinating behavior is due to nanoscopic gold and silver particles embedded in the glass. However, it took 1500 years and doubtlessly countless fantastic interpretations for a plausible explanation to emerge. In 1857, Faraday performed his experiments on metal colloidal solutions. He observed a change in color in dependence of the size and the material of the suspended particles. Just 51 years later, Gustav Mie presented an analytical treatment of the interaction of spherical particles with light. While Mie’s solution was able to explain the size and material dependency of the effects observed by Faraday, the physical processes were not yet fully understood.

Scattering by a small particle

In electromagnetics, the response of a small metallic particle to an external illumination is described by the permittivity ε(λ) of that particle, which depends on the illumination wavelength λ. For coinage metals such as gold, silver, copper, aluminium, the permittivity is complex valued, i.e. has a real and imaginary part, as illustrated for Silver in this figure: the real part ε' in red is usually negative, while the imaginary part ε'' in blue accounts for losses in the material. When the particle permittivity reaches the value ε'=-2, the denominator in the scattering efficiency formula vanishes and the amount of scattered light explodes. Since this phenomenon occurs for a specific wavelength (λ≈350nm in our example), light corresponding to this wavelength is strongly scattered, giving the particle its specific colors. One speaks of a plasmon resonance because the free electrons in the metal behave at this frequency like a solid state plasma in the charged metal ions.

Non-regular shape nanoparticles

The above Mie theory explains the strong colours scattered by small silver nanoparticles in the blue. The experimental reality can however be quite different, as visible in the following dark field microscope image taken by Jack J. Mock from UCSD (data adapted from PNAS vol. 97, p. 996, 2000). While one would expect all silver particles to be blue, some appear yellow, green or even red! Resorting to the high magnification provided by electron microscope on the same sample indicates that the blue particles are spherical and well describes by Mie theory, while the non-regular shape particles shine different colours.

Scattering by non-regular shape particles
Scattering by non-regular nanoparticles

At the turn of the millenium, the NAM has devoted lots of efforts into understanding this surprising phenomenon of the relation between a metallic nanoparticle shape and its colour. The figure here on the left shows the scattering spectra computed for silver nanowires with different cross sections. When the symmetry of the section decreases (i.e. one goes from a circular section to a triangular one), many novel plasmonic modes appear at longer wavelengths. This explains that a low symmetry particle (i.e. a triangle) will have its plasmon resonance in the red, while a high symmetry one will have it in the blue.

The scattering cross section provides information on the colour scattered by the particle in the farfield, i.e. at large distances from the particle. There are many applications of plasmonics however, where the field distribution on the surface of the particle plays a key role. Indeed, each plasmon resonance is associated with a significant field enhancement on the surface of the particle, as illustrated in the following figure. The left part shows the scattering cross section for a 10nm base 20nm perpendicular silver nanowire. Several plasmon resonances are visible in the farfield and correspond to specific colours scattered by the particle. The movie on the right part of the figure shows the field enhancement as a function of the illumination wavelength (indicated in the top left corner). The illumination amplitude is unity and for specific resonances, the electric field amplitude can reach several hundreads (see e.g. the main resonance at λ=460nm). This dramatic field enhancement is the driving force behind surface enhanced Raman scattering (SERS), where the vibration fingerprint of a single molecule can be observed.

Scattering by a non-regular shape particle

Scattering crossection for different illumination directions.


Near-field distribution as a function of the illumination wavelength.


Plasmonic antennas

Over the years, the NAM has explored in detail the optical properties of a broad variety of plasmonic nanostructures. Among those, a geometry appears particularily interesting: it is composed of a pair of metal nanoparticles, separated by a small gap. By analogy with the dipolar antenna fromed by two metallic arms, we call this structure a plasmonic dipole antenna. Typical dimensions for each particle is 70 to 100nm in length with a gap between 10 and 30nm separating both particles. In addition to producing a very strong near-field enhancement, an attractive feature of the plasmonic antenna is that it can be tuned over the entire visible spectrum range by changing its length or the dimension of its gap. The latter plays a key role for the field enhancement, as illustrated in the movie on the left, which shows the field intensity around the antenna for closing gaps.

We are using plasmonic antennas for a broad variety of applications, including fluorescence and Raman spectroscopy, plasmonic trapping and biosensing.