Supramolecular fibers draw widespread attention due to their role in biological systems and ability to form complex materials exhibiting rich and dynamic behavior. Although the information about the supramolecular structure is encoded in their molecular blocks, a complete understanding of how this information translates into the final structure requires detailed insights into the energy landscape of the process and the possible routes across this landscape. Here, we study the formation of 1,3,5-cyclohexanetricarboxamide fibers by a Markov state model of molecular dynamics simulations with the polarizable CHARMM Drude force-field. We provide insights into all stages of supramolecular fiber formation up to microsecond timescales, starting from primary nucleation, through fiber elongation and secondary nucleation, and finally, the assembly of single fibers into bundles. Our results show that nucleation involves a rapid collapse of dissolved monomers into disordered assemblies, which gradually transform into nuclei and then grow into elongated fibers. Moreover, elongation and secondary nucleation appeared to be competing processes, depending on the density of the monomers adsorbed at the fiber-liquid interface. Finally, bundling involves the initial association of fibers by interactions between surface-exposed groups, followed by stabilization of the bundle by surface reorganization, which allows for favorable interactions between aromatic groups.

In recent years, computational methods have become an essential element of studies focusing on the self-assembly process. Although they provide unique insights, they face challenges, from which two are the most often mentioned in the literature: the temporal and spatial scale of the self-assembly. A less often mentioned issue, but not less important, is the choice of the force-field. The repetitive nature of the supramolecular structure results in many similar interactions. Consequently, even a small deviation in these interactions can lead to significant energy differences in the whole structure. However, studies comparing different force-fields for self-assembling systems are scarce. In this article, we compare molecular dynamics simulations for trifold hydrogen-bonded fibers performed with different force-fields, namely GROMOS, CHARMM General Force Field (CGenFF), CHARMM Drude, General Amber Force-Field (GAFF), Martini, and polarized Martini. Briefly, we tested the force-fields by simulating: (i) spontaneous self-assembly (none form a fiber within 500 ns), (ii) stability of the fiber (observed for CHARMM Drude, GAFF, MartiniP), (iii) dimerization (observed for GROMOS, GAFF, and MartiniP), and (iv) oligomerization (observed for CHARMM Drude and MartiniP). This system shows that knowledge of the force-field behavior regarding interactions in oligomer and larger self-assembled structures is crucial for designing efficient simulation protocols for self-assembling systems.

Here, transient supramolecular hydrogels that are formed through simple aging-induced seeded self-assembly of molecular gelators are reported. In the involved molecular self-assembly system, multicomponent gelators are formed from a mixture of precursor molecules and, typically, can spontaneously self-assemble into thermodynamically more stable hydrogels through a multilevel self-sorting process. In the present work, it is surprisingly found that one of the precursor molecules is capable of self-assembling into nano-sized aggregates upon a gentle aging treatment. Importantly, these tiny aggregates can serve as seeds to force the self-assembly of gelators along a kinetically controlled pathway, leading to transient hydrogels that eventually spontaneously convert into thermodynamically more stable hydrogels over time. Such an aging-induced seeded self-assembly process is not only a new route toward synthetic out-of-equilibrium supramolecular systems, but also suggests the necessity of reporting the age of self-assembling building block solutions in other self-assembly systems.

Reported here is a 2D, interfacial microcompartmentalization strategy governed by 3D phase separation. In aqueous polyethylene glycol (PEG) solutions doped with biotinylated polymers, the polymers spontaneously accumulate in the interfacial layer between the oil-surfactant-water interface and the adjacent polymer phase. In aqueous two-phase systems, these polymers first accumulated in the interfacial layer separating two polymer solutions and then selectively migrated to the oil-PEG interfacial layer. By using polymers with varying photopolymerizable groups and crosslinking rates, kinetic control and capture of spatial organisation in a variety of compartmentalized macroscopic structures, without the need of creating barrier layers, was achieved. This selective interfacial accumulation provides an extension of 3D phase separation towards synthetic compartmentalization, and is also relevant for understanding intracellular organisation.

Coarse-grained molecular dynamics simulations are employed to obtain a detailed view of the formation of long-range ordered lamellar structures of physisorbed self-assembling long functionalized alkanes on graphite. During the self-assembly, two processes take place: Langmuir preferential adsorption and rearrangement on the surface. The rearrangement starts with nucleation, in which molecules create an ordered domain. The nucleation mechanism is temperature dependent. At lower temperature independent, small and stable nuclei seed the emergence of long-range ordered domains. In contrast, at a higher temperature, molecules adsorb on the surface, and only when a certain level of surface coverage by the adsorbent is reached, the whole structure undergoes a transition from a liquid-like structure to an ordered structure. After this step, relatively slow corrections of the structure take place by Ostwald ripening.

In this thesis, we explore the application of various molecular simulations techniques to give insights into the self-assembly of supramolecular systems. Chapter 1 explains the importance of molecular simulation to study the selfassembly process. In chapter 2 we study self-assembly of a derivative of 1,3,5-triamidocyclohexane (CTA) using common techniques: simulations of selfassembly from randomly distributed molecules and simulations of the final structure. The results show the importance of the choice of force-field and limitations of conventional molecular dynamics to give insights into processes which occur on a long timescale. In Chapter 3 we tackle the timescale issue by using an adaptive sampling method. The results provide a unique insight into the kinetic pathways of the self-assembly process. Moreover, we were able to provide insights into the next stages of the self-assembly. Although, the method provides insight at a level of detail hardly accessible by any other technique it is limited to rather small systems. In Chapter 4 we study self-assembly of long functionalized alkanes on a graphite flake. We use coarse-grained molecular dynamics to tackle both temporal and spatial scales instead of the high resolution of the all-atomistic model. These results give insights into the mechanism of self-assembly of monolayers on graphite. Chapter 5 is an extension to Chapter 4. Here, we study the last stage of the self-assembly process, Ostwald ripening, responsible for correction of the structure, which leads to high quality long-range ordered assemblies. The results presented in this thesis have two major outcomes: (a) methodology to simulate of self-assembly, and (b) insights into self-assembly process.

Ostwald ripening can improve the long-range order of self-assembled monolayers by the growth of large domains and disassembly of smaller ones. Here, coarse-grained molecular dynamics simulations are used to study the dynamics of the stable assembly and the coarsening of defects of physisorbed monolayers of long-chain functionalized alkanes. Our results show that the partial desorption from the surface of one or more adsorbent molecules is the essential process that allows other adsorbent molecules to rearrange on the surface and thereby improve alignment. We also show that the ripening process is faster at higher temperature because the rate of partial desorption is higher.

We report on the biocatalytic activation of a self-assembling (unprotected) tripeptide to stabilize oil-in-water emulsions on-demand. This is achieved by the conversion of a phosphorylated precursor into a hydrogelator using alkaline phosphatase (AP) as the trigger. The rate of conversion, controlled by the amount of enzyme used, is shown to play a key role in dictating the morphology of the nanofibrous networks produced. When these amphiphilic tripeptides are used in biphasic mixtures, nanofibers are shown to self-assemble not only at the aqueous/organic interface but also throughout the surrounding buffer, thereby stabilizing the oil-in-water droplet dispersions. The use of enzymatic activation of tripeptide emulsions gives rise to enhanced control of the emulsification process because emulsions can be stabilized on-demand by simply adding AP. In addition, control over the emulsion stabilization can be achieved by taking advantage of the kinetics of dephosphorylation and consequent formation of different stabilizing nanofibrous networks at the interface and/or in the aqueous environment. This approach can be attractive for various cosmetic, food, or biomedical applications because both tunability of the tripeptide emulsion stability and on-demand stabilization of emulsions can be achieved.