Search

The shaping of liquids opens a whole new paradigm of applications in various fields such as additive manufacturing, chemical processes, energy storage, and membrane-based applications. In my PhD project, I investigated different methods of shaping liquids. I focused on two approaches, first by external means such as gravity or hydrodynamic forces to deform the equilibrium shape of the liquid and further stabilize the deformed shape with a thin film of particles at the interface. With this approach, we introduced a new facile method of structuring liquids in all liquid systems. Second, I used the internal remodeling of a liquid solution to generate a high surface area liquid mesh of oil and water channels called bijel. My research enhanced our knowledge of the different factors that affect bijel structure and made it possible to control the structure and scaling more effectively. This study also promoted the research on the flow properties of fluids in the liquid channels and particularly the interfacial applications such as catalysis and separations. Finally, To take advantage of the high interfacial area between the liquid phases we introduced particle-based functionality in the bijel as a microreactor. We added catalytic alumina to the liquid interface for carrying out the chemical production of high-value epoxy compounds. To summarize, through my PhD research I used particles based stabilization of various liquid architectures having potential to be the host material for various applications in chemical and energy storage industry.

Iron deficiency is one of the most prevalent nutritional problems in the world. However, iron is a challenging mineral to add to food products. Iron-containing compounds can react with the (phyto)chemicals present in foods and as a result cause severe changes in the organoleptic properties, for instance off-flavor and off-color. To date, iron fortification of foods has proven to be an efficient and cost-effective approach to overcome iron deficiency. Iron-containing compounds that are applied as iron fortificants are divided into three main categories; water-soluble, poorly watersoluble (but soluble in dilute acid), and water-insoluble (and poorly soluble in dilute acid). Although water-soluble compounds have the advantage of high iron bioavailability, the other categories are more in the center of attention because of their minimum influence on the organoleptic properties of the foods which is a consequence of their limited solubilities. Among the poorly water-soluble or water-insoluble iron compounds, ferric pyrophosphate (Fe(III)PP) has attracted a great deal of attention. Fe(III)PP is a white solid that prevents addition of unwanted colors to foods. It has been shown that Fe(III)PP is very poorly soluble in the food relevant pH (3-7) which is the reason for the limited reactivity of this salt with the fortified food vehicle. Furthermore, Fe(III)PP has low solubility at low pH and enhanced dissolution at high pH which is advantageous for ensuring the sufficient iron bio-accessibility. However, it has previously been shown that addition of iron in the form of Fe(III)PP cannot fully prevent the discoloration of the phenolic-rich foods. Therefore, improving the function of Fe(III)PP as an iron fortificant (i.e., decreasing the iron-mediated reactivity while ensuring the iron bio-accessibility) still remains of interest. In this work, we seek the strategies by which we can design iron-containing compounds with minimum solubility in the food-relevant pH (3-7), and high and/or fast dissolution in gastric and intestinal pH (1-3 and 6-8, respectively). Interestingly, mother nature can help us find the answer. Inspired by naturally-occurring minerals such as anapaite (i.e., a mixed calcium–iron phosphate mineral), we intend to embed iron in the matrix of a second (divalent) metal (or mineral) salt, which is less chemically reactive, aiming for: (i) decreasing the iron-mediated reactivity to preserve the organoleptic properties of the food vehicle, and (ii) increasing iron dissolution from the designed multi-mineral salt in the gastric conditions to ensure bio-accessibility of iron (and the other mineral). Another benefit of using these multi-mineral salts is the possibility of simultaneous delivery of at least two minerals by the fortified food vehicle. In Part I of this thesis, we explore the possibilities of improving the function of Fe(III)PP as an iron-fortificant by mixing Ca along Fe in one salt matrix. Part II of the present thesis is dedicated to iron (II)-containing pyrophosphate salts that are potential iron fortificants. In this part we explore the possibility of applying ferrous pyrophosphate (Fe(II)PP) in food fortification. Finally, in Part III of this thesis, we challenge the notion that cooperative binding only happens in complicated biological systems like hemoglobin.

In this thesis we have investigated how the dynamics of particle are affected by surface activity, which is the property of particles to locally alter the solute concentration through for example a surface reactions or dissolution. We found that surface activity can have three effects on the particle dynamics. First, it can cause the particles to self-propel. Surprisingly a heterogeneous surface activity is no prerequisite for this and also particles with isotropic surface activity can swim due to a hydrodynamic instability, provided the activity is larger than a threshold value. Particles that move due to this instability are called isotropic swimmers. In Chapter 2 we studied the swimming dynamics of droplet that slowly dissolve in surfactant solution as a model for such isotropic swimmers. We found that their persistence time can be tuned through droplet size and the surfactant concentration. This finding suggests that stochastic character in the motion of active materials on granular length scales is not only caused by Brownian rotation of these active particles. Rather we think that fluctuations in the fluid flow or spatial inhomogeneities in the dissolution rate cause stochastic turning. Second, we found that even below the onset of swimming, the dynamics of particles with homogeneous surface activity are enhanced or attenuated by the activity, depending on whether solute is consumed or produced. In Chapter 3 we investigated theoretically the instability that gives rise to self-propulsion for isotropic particles. We found that particles with a surface activity just below the swimming threshold can coast as if they were inertial, even though they are in the low Reynolds number regime. We made an attempt to test this finding experimentally, but the results remain inconclusive. Third, surface activity induces effective interactions between particles. We measured such solute-mediated interactions between two dissolving oil droplets in Chapter 4 and found that the interaction scales with inter-particle distance as $1/r^2$. Moreover the interaction strength increases with droplet size and surfactant concentration. Because solute-mediated interactions are dissipative and involve the solvent, they can have the unique property that particle 1 is attracted by particle 2, while particle 2 is repelled by particle 1. This asymmetry in solute-mediated interactions can lead to chemotactic chasing, when the interaction strengths are properly tuned, as we show in Chapter 5. We also show that clusters of chasing droplets can move translationally, rotationally or reorganize depending on their geometry. We made a step in the direction of applying the knowledge of the phenomena that we learned for dissolving droplets to solid colloids. The biggest hurdle standing in the way of that comparison is that the phoretic mobility for solid particles is unknown for many solute gradients and not easy to measure experimentally. In Chapter 6 we first reproduced earlier measurements of diffusiophoretic mobilities of solid particles using microfluidic devices. Then we set out to improve this technique so that it requires fewer particles and no longer relies on the particles being fluorescent.

Colloidal particles are defined as objects with at least one dimension in the size range of approximately a few nanometers to a few micrometers. The analogy between colloids and molecules is applied to bridge the gap between atomic and molecular world to structured materials. A wide range of structures can be self-assembled from building blocks of different shapes, components and functionalities. However, directional forces are required for programmable self-assembly of colloids into complex hierarchical architectures. Directional interactions can be achieved experimentally via chemical patches, the combination of particle shape and depletion forces, or through capillary interaction at the liquid-liquid interface. In this thesis, we aim to develop anisotropic particles that can be used for directional assembly. This thesis is organized in two parts. In the Part 1, consisting of Chapter 2-4, we focus on the synthesis of patchy particles with orthogonal functionality and with anisotropic shapes. First, we develop a new approach based on a modified two-step seeded dispersion polymerization to prepare patchy spheres with orthogonal patches. The resulting particles feature tunable patch ratio, and chemically orthogonal handles for further chemical modification (Chapter 2). Next, we extend the patchy particles to anisotropic shape. In Chapter 3 we report a general route towards anisotropic patchy particles by combining seeded dispersion polymerization and seed mediated heterogeneous nucleation. By varying the hydrophobicity of the seeds, we could achieve anisotropic particle with the shape of snowman, dumbbells and trimers etc. (Chapter 3). Then our attention turns towards another type of colloids, dimpled particles. We present a facile method to prepare dimpled particles using seeded dispersion polymerization and propose a new mechanism to prepare dumbbell shaped particles with a transition from solid dumbbells to dumbbells with a cavity at the protrusion site. (Chapter 4). In Part 2, we transfer patchy particles to liquid-liquid interfaces to investigate how active and passive patchy particles behave at the confined environment. First, we show that the capillary interaction between Janus dumbbells is induced by both shape and chemical heterogeneity of the dumbbells. By controlling the geometry and the difference in surface hydrophobicity between two lobes of the Janus dumbbells, we could vary the configuration at the interface and the evolution of the self-assembled micro-structures (Chapter 5). Finally, we investigate the behavior of active Janus spheres at a two-dimensional liquid-liquid interface with hexagonal polystyrene lattice as obstacles (Chapter 6).

The complexity seen in biological and soft systems often precludes a first principles approach. In order to gain a good understanding of such complex systems, simplification is needed. As systems become larger, the interplay between the underlying mechanisms and details leads to complex system-wide behaviour. Very often this behaviour will be common to a large group of otherwise unrelated systems. Simple model systems can represent a wide variety of such systems, even though the underlying chemistry is different. Such toy models are immensely instructive in the understanding of more complex systems. In this thesis we describe and analyse toy models for two different soft and living systems. In the first part we consider an experimental model system that is representative of a broad class of materials consisting of ordered membranes. We study the self-assembly of a mixture of SDS and β-cyclodextrin (β-CD) into concentric hollow microtubes using small-angle x-ray scattering. After a concentration dependent waiting time we observe the appearance of hollow cylinders that grow inward. The distribution of waiting times follows the non-linear scaling with SDS/ β-CD concentration that is predicted by classical nucleation theory for a two-dimensional critical nucleus. Moreover, when the experimental time is rescaled according to classical nucleation theory, the entire trajectory of inward growth collapses onto a single curve, indicating that a nucleation process determines the entire kinetics of inward growth. The mechanism of inward growth can therefore be explained by the successive nucleation of new, discrete cylinders inside previous existing ones, constricted in their size by the size of the original tube. In the second part we build upon a toy model based on equilibrium binding of ligands to a template, as a model for transcription regulation. Transcription initiation is a complex process involving multiple steps, which, in its most simplified form, can be described in three steps: (1) binding of RNAP to the promoter, (2) (irreversible) isomerization to an open complex, followed by (3) escape of the open complex to form an RNAP complex active in transcription. When the rearrangement of RNAP and transcription factors is fast compared to the formation of an open complex, we can assume that the rate at which the open complex is formed – the first kinetically significant step in the transcription process – is proportional to the occupation probability of the promoter by RNAP. Thermodynamic theory, based on the toy model of ligand adsorption to a template, has been developed to calculate this probability. Such thermodynamic models are traditionally derived in the limit of genes in isolation, within a canonical ensemble. However, individual regulatory proteins are typically charged with the simultaneous regulation of a battery of different genes. As a result, when one of these proteins is limiting, competitive effects have a significant impact on the transcriptional response of the regulated genes. We present a general framework for the analysis of any generic regulatory architecture that accounts for the competitive effects of the regulatory environment by isolating these effects into an effective concentration parameter.

The main goal of this thesis is to increase our understanding of colloidal self-assembly processes and develop new strategies to assemble colloidal building blocks into more sophisticated and well-defined super-structures. Self-assembly is a spontaneous process in which a disordered system of pre-existing building blocks forms an ordered structure without human intervention. For example, virus capsid proteins can self-assemble into virus microcapsules. However, direct investigation of the self-assembly process of, for examples, proteins and other molecules in situ, is difficult since those objects are too small and move too fast to be tracked directly by techniques such as electron and optical microscopy. Herein, we employ colloids as models of (macro) molecules to study self-assembly. This thesis divides into two parts. In the first part, we show various methods to tune the properties of the colloids, including shape, charges, morphology, size and surface properties. In the second part, we focus on the self-assembly of spherical colloids into one, two and three-dimensional colloidal aggregates using different principles. We show that the delicate balance of short-range hydrophobic attraction and relatively longer-range electrostatic repulsion can result in the formation of a Bernal spiral-like structure. While the use of a good solvent of colloids during the vertical deposition process allows for the formation of floating colloidal crystal monolayers. We also show that pH reversible encapsulation of oppositely charged colloids with a vast size difference can be achieved in the presence of pH-responsive polyelectrolytes in solution.

Self-assembly is the spontaneous formation of patterns or structures without human intervention. This thesis aims to increase our understanding of self-assembly. In self-assembly of proteins, the building blocks are very small and complex. Consequently, grasping the basic principles that drive the formation of, for example, viruses is challenging. Colloidal particles, on the other hand, are much larger and can be used as relatively simple model particles. In this thesis, we investigate how deformability and near-critical phenomena can direct self-assembly and the formation of well-defined materials. The results in this thesis show that deformability and near-critical phenomena can take colloidal self-assembly a significant step further. Deformability allows relatively simple building blocks to self-assemble into complex structures. This enabled the first colloidal realisation of self-assembly into microcapsules without a template. Our simple building blocks are mutual attractive, anisotropic and deformable. These characteristics can also be recognized in building blocks of viruses, suggesting that these could be important in the self-assembly of virus microcapsules as well. Near-critical phenomena, on the other hand, allow for externally tunable, directional interactions. Consequently, anisotropic particles thermoreversibly self-assemble into structures that resemble micelles of molecular surfactants. We expect that externally tunable, directional interactions could facilitate self-assembly into other structures that are up till now kinetically inaccessible. Near-critical phenomena also result in thermoreversible attachment of colloids onto the surface of phase-separated droplets. We demonstrate that thermoreversible attachment can be applied to prepare and destabilize emulsions by simply changing the temperature.

The formation of structures from a disordered starting point is a common feature in biological systems. Examples include formation of virus capsids from proteins, folding of proteins and formation of micro-tubules. Capturing the physics behind these self-assembly processes is not a trivial task due to two fundamental reasons. Firstly, the individual building blocks are of molecular dimensions and therefore not easily visualized. The second difficulty arises from the fact that the interactions between the building blocks are complex and compose of various contributions i.e. electro-statics, hydrophobic interactions, steric repulsion and hydrogen bonding. The importance of each individual contribution is generally difficult to extract from experimental results, since independent variation of contributions is usually not possible. In this context, the use of colloidal particles as analogues for the molecular building blocks is an attractive option. Colloids are objects with at least one dimension in the order of 1 nm – 1 μm, that are dispersed in a continuous medium. Thermal fluctuations result in constant uncorrelated movements of the colloids, similar to the motion of molecules and atoms. However, due to the relatively large dimensions of colloids compared to atoms and molecules, the dynamics are much slower. The large size and related slow dynamics facilitate direct visualization and therefore studying structure formation upon assembly of colloidal building blocks. An additional advantage of using colloidal systems is that inter-particle interactions and particle shape are tunable to a large extent. Unlike molecular systems, colloidal interactions are tunable in magnitude and range by physical/chemical alternations. This allows for the rational design of particles to either mimic molecular self-assembly or the formation of new assemblies with tailored properties. Systematic variation of either shape or interactions can in principle reveal the minimal complexity required to form a desired superstructure. The focus of this thesis can be divided into two main topics. In this first part we explore synthesis procedures for the preparation of large quantities of anisotropic, dumbbell-shaped colloids. These particles consist of two lobes of which one bears chemically modifiable moieties. In principle this allows for introducing directional interactions between the particles. Therefore, these particles are promising candidates as building blocks in self-assembly studies. In the second part we exploit chemical surface modifications of colloids to tune the inter-particles interactions. We mainly focus on Atom Transfer Radical Polymerization as the modification tool. With this technique we are able tether well-defined polymers on the colloidal surface. The physical properties of the immobilized polymers are used to steer the behavior of the resulting colloidal system. This enables us to prepare colloidal systems that showed thermo-reversible aggregation and the first colloidal system that shows out-of-equilibrium/dissipative assembly. An alternative modification strategy relies on controlling the overall charge of colloidal particles by performing a surface reaction that introduces positive moieties on a negative charged particle. The extend of the reaction determines the ratio between positive and negative groups and therefore the overall charge of the particles. These particles are subsequently used in electrostatic driven assembly to form well-defined and finite-sized colloidal aggregates.

In this thesis the self-assembly of anisotropic polystyrene colloidal particles is studied using optical microscopy. These particles consist of different lobes with attractive and non-attractive interactions. This anisotropy in inter particle interaction is induced by depletion attraction combined with a difference in surface roughness between the lobes. The shape of the particles that are used as building blocks has a profound effect on the structures formed by self-assembly. Snowman or dumbbell-shaped particles consisting of one attractive (smooth) and one non-attractive (rough) lobe self-assemble into spherical micelle-like structures. These particles can also be used to encapsulate and stabilize larger spherical particles. Triangular particles on the other hand, consisting of one attractive and two non-attractive lobes, resembling a “Mickey Mouse” head, self-assemble into elongated tube-like structures. These structures are observed with optical microscopy in the experimental system and supported by Monte Carlo simulation results. Understanding this effect of building block shape on the resulting structure is important for the design of building blocks for the formation of new, functional structures by self-assembly. These structures could for instance be used as vehicles for targeted drug delivery. The geometry of dumbbell-shaped particles also has an effect on the crystalline ordering of these particles by convective assembly. A larger particle length (less overlap between the lobes) results in reduced crystalline order, while crystals of these particles have interesting optical properties with possible application as photonic crystals.

The central theme of this thesis is the interplay between colloids and interfaces. The adsorption of colloids at fluid-fluid interfaces is the main topic and covers Chapters 2-6. Pickering emulsions where colloidal particles act as emulsion stabilizers in the absence of surfactants are studied in a number of systems with different colloids, particle shapes and oils. Interfacial particle adsorption is widely employed in food, cosmetic and pharmaceutical applications and plays a pivotal role in oil recovery and metallurgical refining. The conditions under which Pickering emulsions may form spontaneously are discussed in Chapter 2. The spontaneous formation and the stability of the Pickering emulsions of varying composition are achieved by a collective effect of solid particles, amphiphilic ions and interfacial tensions of the bare oil-water interface of ~10 mN/m or below. The observed stability and formation of emulsions of different composition point to a new class of solid-stabilized meso-emulsions. The effect of particle shape anisotropy in the stabilization of emulsions is investigated in Chapter 3 by employing for the first time cubic, ellipsoidal and peanut-type hematite microparticles. The interfacial packing and orientation of these anisotropic microparticles are revealed at the single particle level by direct microscopic observations. Emulsions are stable against further coalescence for at least one year. The creation of surfactant-free Pickering foams with anisotropic hematite microparticles and their dependence on ionic strength are studied in Chapter 4. The microparticles cover bubbles by densely packed interfacial monolayers that provide high stability against disproportionation and coalescence. Moreover, solid-stabilized air bubbles are used as scaffolds for the creation of free-standing particle films that are, in fact, inorganic bilayers that only consist of cubes. In Chapter 5 the self-assembly of colloidal silica cubes at oil-water interfaces is discussed. A combination of optical and laser scanning confocal microscopy enables the in-situ study of both the packing as well as the orientation of the colloidal cubes at the interface. Single layers of cubic particles arrange in ordered domains, displaying a packing intermediate between cubic and hexagonal. Moreover, the cubes show a preference for orienting parallel with the interface. Solid-stabilized emulsions from natural resources are presented in Chapter 6 by using colloids, prepared from the water-insoluble corn protein zein, and soy bean oil. Zein colloids are synthesized via an anti-solvent precipitation procedure and employed in the formation of stable oil-in-water Pickering emulsions as a function of particle concentration, pH and ionic strength. Finally, a thermo-reversible colloid-in-tube co-assembly approach that couples molecular self-assembly with colloidal self-assembly is introduced in Chapter 7. While surfactant and cyclodextrin molecules form microtubes, colloids assemble into a library of dynamic colloidal structures within those microtubes. Isotropic spheres form straight, zigzag, and zipper chains depending on the tube-sphere size ratio. Double and triple helical structures, a common occurrence in nature, were even generated from colloidal spheres. Moreover, we demonstrate that the co-assembly of microtubes and colloids is generic for colloids with different shapes and materials. The hierarchical colloid-in-tube co-assembly provides a novel route to temperature-sensitive particle alignment and their release near human-body temperature.

The objective of this thesis work was to incorporate (mixed) dietary minerals into a colloidal, colourless, inorganic matrix material suitable for food fortification, as food fortification is often problematic due to the interaction of ionic minerals with the foodstuffs to which they are added. In search of a stable colloidal dispersion of such a material, we have used various methods to prepare metal pyrophosphate salts containing divalent mineral cations (MIIPPi), trivalent cations (MIIIPPi) or mixtures thereof. We have studied the colloidal properties of the resulting systems and tested the stable dispersions for their reactivity with foodstuffs using a model system

Self-assembly in nature has many facets and occurs on all length scales. This thesis deals with selected examples on the colloidal length scale, including solid-stabilized emulsions, tobacco mosaic virus particles as well as colloidal and colloidal micelles. Part 1 of this thesis deals with the synthesis of anisotropic model colloids, utilizing self-assembly processes. Analogous to atoms that bind to form molecules, colloids with liquid patches are shown to bind with each other forming

Assemblies of Polyoxometalates Polyoxometalates, or POMs for short, come in a variety of shapes and sizes. Some of them are among the largest inorganic molecules known today. These molecules can be highly symmetrical and, as the name already implies, consist of (mainly) metal (molybdenum, tungsten, vanadium, iron, etc) and oxygen atoms in the form of metal oxide polyhedrons. Although individual POM molecules are in itself already a very challenging field of research, assemblies of POMs have great challenges of their own. Take for instance the solution behavior of a subclass of POMs known as Keplerates. Keplerates are highly symmetric, hollow, spherically shaped POMs. In solution they spontaneously assemble into large, hollow, spherical superstructures or POM-shells. These POM-shells have an average radius in water of several tens of nanometer and are composed of a monolayer of more than a thousand of individual POMs. Although the underlying principles behind this spontaneous assembly are still not precisely known, much can already be understood about this curious phenomenon. For instance, the equilibrium size of POM-shells of Keplerate type POMs is inversely proportional to the relative dielectric constant of the medium in which they are dispersed. This behavior is in line with predictions made by a stabilization mechanism based on Coulomb repulsion combined with charge regulation. This model also explains a new structural instability of the POM-shells. Besides the colloidal instability, i.e. the formation of aggregates that consist of many single layered shells, these systems also display an instability on a structural scale within the shell-like assemblies when the ionic strength reaches a critical value. Not only the ionic strength has an influence on the properties of POM-shells in solution. It has been claimed that the number of charges on the POMs themselves plays an important roll in their formation and stabilization as well. For instance, for the Keplerate {Mo72Fe30} experiments show that the interaction energy between the POMs in the shell becomes stronger when the charge density on the POMs increases. This interaction energy can also be estimated from analysis based on the charge regulation model in combination with a model for defects on a sphere. The charge regulation model therefore gives much insight into the behavior of the POM-shells. Up till now the POM-shells have been considered thermodynamically stable. Our experimental results however, challenge this view. It is shown that POM-shells change over time and that the preparation route of solutions of POMs affects the resulting species in solution. Moreover, in concentrated samples, a transition from spherical objects to elongated agglomerates was observed. The elongated objects subsequently grow into large, crystalline, needle-like structures. From these observations it is concluded that POMs follow an unusual nucleation route in which the POM-shells are in fact meta-stable intermediates. The term assembly can also be taken in a more general way. For instance polyoxomolybdates can be used to create extended structures. By using elementary molybdenum oxide building blocks and urea in a one-pot directed synthesis, an inorganic-organic hybrid extended structure, consisting of right- and left-handed helical units can be obtained.

This thesis reports results from a study of particle dynamics in colloidal glasses in the absence and presence of a gravitational field. It also investigates the reentrant glass transition. All the experimental results are obtained after a real space analysis using a confocal scanning laser microscopy as experimental technique that significantly widens the time window over which colloidal systems can be studied. A fluorescent recovery of bleached regions in concentrated suspensions of fluorescent colloidal hard spheres was followed in real space. This method provided data for mean squared particle displacements up to time scales that are three orders of magnitude beyond those available by present experimental techniques like dynamic or static light scattering. It was shown that, above the (hard sphere) glass transition density, particles move over distances on the order of their own diameter on time scales of 106 to 108 Brownian times. Moreover, the mean squared displacement, , showed power-law behavior over seven time (?) decades: ??^ (0.30?0.05). This behavior is different from earlier observations by dynamic light scattering. It was argued that these differences are caused by gravity effects. Further on, the influence of gravity on the long-time behavior of the mean squared displacement in glasses of colloidal hard spheres was studied. For the first time, a significant influence of gravity on the mean squared displacements of the particles was presented. In particular, the systems which are glasses under gravity (with a gravitational length on the order of tens of micrometers) showed anomalous diffusion over several decades in time if the gravitational length is increased by an order of magnitude. No influence of gravity was observed in systems below the glass transition density. It was shown that this behavior is caused by gravity dramatically accelerating aging in colloidal hard sphere glasses. This behavior explained the observation that colloidal hard sphere systems which are a glass on earth rapidly crystallize in space. A quantitative analysis of the structure and dynamics of concentrated suspensions of colloids in which the magnitude of the short range attractive potential was increased by adding non-adsorbing polymers was done. These systems undergo a reentrant glass transition upon increasing polymer concentration. The melting of the glass is accompanied by significant changes in the displacement distribution and its moments. However, no significant variations were detected in the shapes of the displacement distributions. Moreover, structural correlation functions and the magnitude of local density fluctuations did not vary significantly between the glass states and the fluid. The influence of gravity on long-time diffusion in glasses was also studied using a new experimental setup that allowed a real space determination of particle displacements in directions parallel and perpendicular to the gravity field. Dispersions of particles were prepared in different solvent combinations that provide different gravitational lengths. It was shown that the direction of the external gravitational field couples to the particle dynamics and influences the values of mean square particle displacements.

This thesis is made up of several studies of boundaries occurring in colloidal hard sphere crystals and phase separated colloid-polymer mixtures. These boundaries can be studied on the particle level, in real space and in real time by confocal microscopy. A general introduction on the experimental systems and on confocal microscopy is given in Chapter 1. This thesis consists of four main parts. The first part deals with stacking disorder in hard sphere crystals. The second part deals with geometrically frustrated hard sphere crystals. The third part describes the statistics of Brownian interface fluctuations in phase-separated colloid-polymer mixtures. The final part of this thesis deals with the transport of particles through such an interface. In the first part of this thesis we study stacking disorder in hard sphere crystals. Due to the small difference in free energy between the hexagonal close packed (HCP) and face centered cubic (FCC) crystal configurations, both structures can coexist within these crystals. As a result the crystals are often made up of a random combination of ABA and ABC type stacking sequences of hexagonal layers. Chapters 2 and 3 describe in-plane stacking disorder in hard sphere crystals. Here, we show that stacking can change as well within a hexagonal layer, both via abrupt transitions through line-defects (Chapter 2) and via continuous transitions through lattice deformations (Chapter 3). Therefore the hexagonal layers are made up of multiple islands of FCC or HCP stacking type. In Chapter 3 the relation between lattice vacancies and stacking transitions is investigated as well. Due to the osmotic pressure imbalance, particles are pushed slightly toward the vacancy. Furthermore, in areas where lattice deformations occur due to stacking transitions, the vacancy concentration is much higher, allowing the accommodation of lattice deformations. Chapter 4 describes the relation between grain size and typical stacking island dimension. Using methods to distinguish both the stacking type and the stacking direction, we are able to identify the typical island dimension. The fraction of FCC type particles determines the relative size of FCC and HCP type islands, which we can relate to lateral islands with an A, B or C type lateral position through simulations. The second part of this thesis describes how geometry affects hard sphere crystals and crystallization. Chapter 5 describes the crystallization of hard spheres near large, nearly immovable objects: much larger spheres, which are in a way the simplest conceivable impurities or dopants. Crystal nucleation, the initial formation of sufficiently large crystals may be facilitated if the curvature of the dopant is sufficiently low. The subsequent crystal growth of these small crystals is retarded by the presence of impurities, but due to their high curvature, small impurities slow the crystal growth much more down than larger ones. Chapter 6 deals with the obtained crystal structure after crystallization has completed. We introduce a frustration length which quantifies the extent of lattice distortion depending on the size of the dopants. If the dopants are sufficiently close, we show that grain boundaries may form in between directly, which is a result of the delayed growth process in between the impurities. Whether the frustration lengths persist in the sample or slowly anneal out on time depends on the impurity spacing. Chapter 7 compares lattice frustration by two methods: 1) the insertion of impurities and 2) a polygonal particle shape. Both sources of lattice frustration induce polycrystallinity. The third part of this thesis concerns the liquid-liquid interface of phase separated colloid polymer mixtures. Such interfaces are characterised by large interface fluctuations due to the low interfacial tension. Experiments and theory on the residence time of a certain fluctuation above a certain height and the related waiting time in between such heights are presented in Chapter 8. In the final part of this thesis the transport of rigid spheres through such a fluctuating interface is presented. In Chapter 9 we compare the approach of droplets and spheres to a deformable interface. The similarity of the problems shows that the approach of the sphere already captures much of the physics of the more complicated approach of a droplet. The transport of rigid spheres through interfaces is the subject of Chapters 10 and 11. We distinguish between low Bond numbers (Chapter 10) and high bond numbers (Chapter 11), which respectively represent the limits of dominant interfacial tension and gravity. The low Bond number transport configuration is characterized by a draining film in between the sphere and the interface. In the meantime the interface deforms up to maximally about half the sphere's diameter. When the sphere is sufficiently close, it is wetted by the interface and dragged through it, before it starts sedimenting away again. The high Bond number transport configuration is characterized by a thin film of gas which persists around the sphere, while the interface deforms more than half a sphere's diameter. Depending on the Bond number, two scenarios may be observed. At moderate Bond numbers the sphere leaves a V-shaped interface behind. At even higher Bond numbers, the sphere drags a column of material of the phase it was originally in behind it that eventually breaks up through a Rayleigh-Plateau instability.