With the introduction of new excitation/detection schemes in microscopy, such as confocal, multiphoton and total internal reflection, and the advances in fluorophore technology, fluorescence microscopy is today the most rapidly growing light microscopy technique, both in medical and biological sciences. In-vivo imaging with high lateral and axial resolution and large penetration depth can be obtained with confocal microscopy [1]. This technique is selecting the signal from a slice of the object around the focal plane by using a pinhole in the detection that attenuates out-of-focus light. However, confocal microscopy is exciting a large sample volume, and detecting only a small fraction of it (the focal volume), thus creating significantly more photoexcitation than needed. This problem can be overcome by using two-photon (i.e. non-linear) fluorescence microscopy in which two photons with a wavelength in near-infrared region are absorbed [2]. Significant two-photon absorption is reached only in the focal volume where high photon densities are present, allowing for an intrinsic 3D spatial resolution. Moreover, near infrared light is less scattered than the visible light, with consequent improvement of the image contrast and penetration depth.
 
Besides its many advantages, fluorescence microscopy has some limitations. Background fluorescence originates from endogenous object constituents (autofluorescence) or from unbound or non-specifically bound fluorescent probes (reagent background), and  fluorophores under light excitation are prone to irreversible degradation (photobleaching).
 
Recently, colloidal quantum dots (CQDs) have been demonstrated as innovative fluorophores because of their unique optical properties [3]. CQDs are semiconductors nanocrystals (e.g. CdSe core / ZnS shell) of a few nanometers in diameter. The wavelength of the emitted light depends on their size and shape that can be controlled during their synthesis (see Fig. 1) [4]. They show an increasing absorption towards shorter wavelengths, resulting in a broad absorption spectrum, in stark contrast to organic dyes, which feature an absorption peaked about 50-100nm blueshifted to the emission, strongly decaying twoards shorter wavelengths. It is therefore much more practical to excite CQDs emitting at different wavelength by a single monochromatic light source, simplifying multiple marker imaging (Fig. 2). Furthermore, the emission spectrum of CQDs is narrower than that of dyes, enabling an improved spectral discrimination of the emission from multiple CQD markers.
CQDs are also attractive for two-photon fluorescence microscopy due to their high two-photon absorption cross-section. CQDs exhibit high photostability as compared to organic dyes, however surface defects in the crystal structure can act as temporary traps for carriers, preventing their radiative recombination. Furthermore, charging of the CQDs can significantly reduce their quantum yield, The temporal sequence of charging and trapping events often results in intermittent fluorescence (photoblinking) [5], a limitation for CQDs application in fluorescence microscopy.
 
Beyond fluorescence, we can use CQDs excited in resonance it fundamental absoprtion peak to create a strong non-linear signal called four-wave mixing (FWM) which is due to its third-order non-linearity [6]. The aim of this project is to study the applicability of such resonant FWM signal from CQDs as biomarkers both in-vitro and in-vivo, to obtain a novel type of multi-photon imaging, with intrinsic sectioning capability without the need of a detection pinhole to reject out of focus light, alternative to two-photon fluorescence microscopy. Since FWM is a coherent signal directly linked to the third-order polarization, it has the benefit of detection free from any incoherent fluorescence background. Moreover, the spatial resolution can be increased beyond the diffraction limit due to the optical non-linearity.
 

Project Aims

• Implementation of a set-up to measure and image the transient resonant FWM on biocompatible CQDs in solution with sub-micron three-dimensional spatial resolution.
• Quantitative evaluation of the sensitivity of the technique and the related CQD photostability by investigating the FWM signal intensity and long-term stability versus CQD density and excitation power.
• Determination of the lateral and axial spatial resolution of the technique by imaging of reference structures such as patterned arrays of CQDs in polymer matrices.
• Imaging of cells (both fixed and living) labelled with CQDs to assess the applicability of the technique to cell biology.
• Demonstration of zero-background imaging performance by the implementation of a three-pulse FWM.