Optical characterisation of subwavelength dielectric particles using particle tracking beyond the Stokes-Einstein relation

Licentiate thesis, 2021

As the importance of nanoparticles continues to increase in both biology and industrial processes, so does the need for accurate and versatile characterisation methods. However, most light-based methods to quantify size and refractive index of individual particles are either limited to snapshot observations, particles larger than the wavelength of light, non-dynamic particle properties, or assuming the hydrodynamic boundary conditions without experimental evaluation. The aim of this thesis is to partially overcome these limitations by further developing two different characterisation methods based on optical microscopy combined with particle tracking, where the analysis goes beyond the ordinary Stokes-Einstein relation.

The first method combines off-axis holographic nanoparticle tracking with deep learning (Paper I). By utilizing the optical signal, both size and refractive index of individual particles with a minimum size of R=150 nm were accurately determined using only five particle observations. The method was evaluated using particles of different sizes, refractive indices, surrounding media as well as for polystyrene nanoparticle clusters, for which reversible fluctuations of the number of monomers could be resolved while the fractal dimension remained constant.

The second method is based particles tethered to a laterally fluid supported lipid bilayer and quantification of their diffusivity and flow-induced motion (Paper II). By separating the friction contributions from the tethers and the particle, simultaneous measurement of size and diffusivity enabled a comparison with theory using partial slip as a fitting parameter. This was used to quantify the slip length for different lipid vesicles, and to clarify the size-dependent mechanistic aspects concerning the mobility of membrane-attached nanoparticles.