The demand for ever more complex nanostructures calls not only for new design concepts and synthetic protocols, but also incorporation of more advanced characterization methods and more refined data analysis tools.
In this thesis, challenges to characterize complex molecular and colloidal self-assemblies are explored using model systems. In Publication I we study the molecular self-assembly of polyelectrolyte complexes into multicompartment micelles with intricate ''turbine-like'' surface morphologies. We demonstrate how conventional 2D transmission electron microscopy (TEM) is not sufficient to resolve such morphologies and thus 3D reconstruction based on electron tomography (ET) is needed. In Publication II we show how ET allows resolving chiral templated ionic self-assemblies of gold nanoparticles on cellulose nanocrystals. In Publication III, ET allows resolving capsid-like hollow superstructures of cobalt nanoparticles driven by hydrogen bonding between the nanoparticle ligands. In Publications III and IV we also investigate how molecular simulations can aid the interpretation of the experimental findings. Publication IV discusses structural characterization of pentablock quintopolymeric vesicular superstructures in solvent media. In simpler block copolymers, various staining methods allow the identification of the self-assembled domains. In self-assemblies involving 5 different microdomains resolving the microdomains and their structures was a real challenge. We implemented advanced inversion methods such as total variation in ET to differentiate between the different blocks of the self-assemblies.
We discuss in detail the nature and mathematics of forward and inverse problems, including the specific cases ET and dynamic light scattering (DLS). We also show how the use of conventional spectroscopic methods, such as DLS and its data analysis routines, can result in fallacious conclusions when applied to complex self-assemblies, thus demonstrating the need for more diverse data collection in combination with Bayesian analysis for reliable inference.
The investigations of this thesis pave way towards more advanced characterization methods and data analysis in electron microscopy and light scattering, which we foresee to be increasingly required for understanding the future complex self-assemblies.
|Publication status||Published - 2018|
|MoE publication type||G5 Doctoral dissertation (article)|
- block copolymers, cellulose nanocrystals, metal nanoparticles, micelles, vesicles, dynamic light scattering, transmission electron microscopy, tomograph, Bayesian inference, inverse problems