Mid-infrared properties of heavily-doped semiconductor nanocrystals – MOSAIC

Metasurfaces based on plasmonic nanoparticles, which collect and concentrate light into subwavelength volumes, have enabled a whole new family of planar optical devices including light sources, energy harvesting systems, light wavefront engineering components, non-linear systems, biosensors and photocatalysis devices. Indeed, from a technological perspective, they offer several advantages over pure dielectric devices: reduced pixel area, sub-wavelength resolution and locally large field enhancement leading to low power consumption, high amplitude signals and local heating.
The first successful plasmonic metasurfaces were made out of noble metals and patterned using lithography techniques. The situation might be about to change, however. First, in order to pattern the structures, it has been recently proposed to use chemically synthesized, randomly deposited metallic nanoparticles instead of the expensive top-down lithography techniques. Second, new plasmonic materials have recently emerged and open radically new possibilities. Heavily-doped semiconductors, in particular, have been put forward for two reasons: their lower carrier density compared to noble metals shifts the plasma frequency towards the infrared, where the applications are numerous but the devices are scarce and expensive; in addition, their optical properties can be tuned over a large range of wavelengths by dynamic free carrier modulation.
Combining the advantages of chemically synthesized nanoparticles and new plasmonic materials, colloidal heavily-doped semiconductor nanocrystals may thus revolutionize the design of plasmonic metasurfaces. In order to fully exploit their potential, however, a leap in our understanding of their optical properties is required. A few recent experiments have provided first sets of data, but most of them suffer from a crippling limitation: they were performed on nanocrystal ensembles and could not disentangle the intrinsic properties of individual nanocrystals from statistical effects due to the nanocrystal heterogeneity, or from collective effects due to optical coupling between the particles. The only notable exception is a spectroscopy measurement of the optical response of individual ZnO:Al nanocrystals realized with a synchrotron infrared source coupled to a nanoscale Fourier transform infrared spectroscopy (nano-FTIR), a technique that can obviously not be applied systematically owing to the requirement of accessing a large facility.
The objective of the MOSAIC project is to solve these issues by using a unique combination of state-of-the-art experiments and numerical simulations. We will first build a table-top, high-sensitivity, background-free nano-FTIR experiment to directly measure the complex polarizability of individual nanocrystals. By combining these measurements with advanced numerical calculations, we will be able to precisely model the intrinsic dielectric permittivity of several families of heavily-doped semiconductor nanocrystals beyond the classical Drude model. We will then extend our study to ensembles of nanocrystals using classical FTIR spectroscopy for the experiment and a multiple-scattering code that we recently developed for the simulations. This code has the unique capability of describing accurately the collective scattering of complex arrangements of particles, even in the presence of a substrate. Going beyond the local Drude model, it also takes into account the size-dependent effects appearing in nanoparticles when the electron gas is confined to a volume much smaller than the wavelength of light to the cube.
Our original approach, breeding theory and experiment, is expected to have a strong impact by providing researchers and industrial actors with the fundamental knowledge required to engineer the optical properties of heavily-doped semiconductor metasurfaces, and open the door to a new generation of dynamical plasmonic devices.