Supported lipid bilayers (SLBs) are unique model systems for biological membranes. SLBs can be formed by fusing liposomes on solid substrates, which can be characterized by a variety of surface analytical techniques, such as Atomic Force Microscopy (AFM), X-ray diffraction, Quartz Crystal Microbalance (QCM), etc. In this study we used AFM to investigate the dynamic process of the formation of SLBs from liposomes in solutions containing metal ions and phase separation between different lipids as a function of temperature. Divalent cations, Ni2+ in particular, was found to be critical to the deposition of bilayers. Lipid rafts are plasma membrane microdomains rich in sphingolipid and cholesterol forming a liquid ordered phase surrounded by a liquid disordered phase. Lipid rafts are insoluble in cold non-ionic detergents, also called Detergent Resistant Membranes (DRMs). The interaction behaviors between detergent (Triton X-100) and mixed bilayers (DOPC/DPPC and DOPC/SpM) were studied by AFM. The way lipid bilayers were solubilized by Triton X-100 was quite different below and above its critical micelle concentration (CMC), and the SpM domains were found to be resistant to detergent extraction in the cold.

In this research, two major experiments, including the self-assembly of silica spheres, were performed by using a physical confinement method with an attractive capillary force. The silica spheres were dragged and aggregated as results of the evaporation of the solvent. In the first experiment, silica spheres were assembled into the two-dimensional pattered substrate, constructed by the photo-resist film formed under a lithography process. Several patterned substrates could work as a physical trap during the flow of the silica spheres. The ordered arrangement of the silica spheres was controlled by the concentration and the size of the silica spheres, the thickness of the photo-resist film, and the titled angle of the substrate. In our conditions, the silica spheres could orderly arrange in larger area of the substrate. In the second experiment, a surface-enhanced Raman scattering (SERS) enhancement was observed from a chain of silica spheres with silver nanoparticles, which worked as a excitation source to provid a strong local electromagnetic fields exciting the crystal violet (CV) dye coated on the silica spheres. We found that the CV molecules has a strong SERS intensity due to the refraction and reflection of the incident light within the silica spheres. When the silica spheres were linearly arranged, longer length of the chained silica spheres would lead to a maximum value of the SERS intensity.

Solubility-based membrane separation, in which the more soluble species preferentially permeates across the membrane, has attracted considerable recent attention due to both economic and environmental concerns. This solubility-selective mode is particularly attractive over a diffusivity-selective mode in applications in which the heavier species are present in dilute concentrations. Examples include the recovery of volatile organic components (VOCs) from effluent streams and the removal of higher hydrocarbons from natural gas. Recently, nanocomposites have shown great promise as possible membrane materials for solubility-selective separations. The chemical derivatization of inorganic mesoporous substrates has been explored to synthesize organic-inorganic nanocomposite membranes. The most exciting feature of this approach is that it enables the rational engineering of membrane nano-architecture with independent control over the free volume and chemistry to create membranes with highly customizable permselectivity properties. In this study, we synthesized the organic-inorganic nanocomposite membranes by decorating the surfaces of commercially available mesoporous alumina substrates, and surfactant-templated highly ordered mesoporous silicate thin films placed on commercially available macroporous inorganic substrates, with a selective organic material that is physically or chemically anchored to the porous surfaces. Hyperbranched melamine-based dendrimers, with nanometer dimension and chemical composition designed to target certain components, were used as filling agents. We evaluated these membranes for several environmentally relevant separations, such as the recovery of the higher hydrocarbon from air and the removal of trace VOCs from air or water, while exploring the impact of organic oligomer size, chemistry, and surface coverage, as well as substrate pore size and structure, on membrane performance. First, we did a model study to verify the feasibility of dendrimer growth inside mesopores by using ordered mesoporous silica. Alumina-ordered mesoporous silica (alumina-OMS) hybrid membranes were prepared as new inorganic porous substrates. Finally, we synthesized dendrimer-ceramic nanocomposite membranes by growing several generations of melamine-based dendrimers with diverse functional groups directly off the commercial alumina membranes. Composite membranes show very high propane/nitrogen selectivity up to 70.

Over the past few decades there has been great interest in exploring alternatives to conventional separation methods due to their high cost and energy requirements. Membranes offer a potentially attractive alternative as they potentially address both of these points. The overarching theme of this dissertation is to design nanocomposite membranes for processes where existing separation schemes are inadequate. This dissertation focuses on three challenges: 1) designing organic-inorganic hybrid membranes for reverse-selective removal of alkanes from light gases, 2) defect-free inorganic nanocomposite membranes that have uniform pores, and 3) nanocomposite membranes for minimizing protein fouling in microfiltration applications. Reverse-selective gas separations that preferentially permeate larger/heavier molecular species based on their greater solubility have attracted considerable recent attention due to both economic and environmental concerns. In this study, dendrimer-ceramic hybrid membranes showed exceptionally high propane/nitrogen selectivities. This result was ascribed to the presence of stable residual solvent that affects the solubility of hydrocarbon species. Mesoporous silica-ceramic nanocomposite membranes have been fabricated to provide defectless mesoporous membranes. As mesoporous silica is iteratively synthesized in the ceramic macropores, the coating method and the surfactant removal step significantly affected permeance and selectivity. It was also shown that support layers can cause a lower selectivity than Knudsen limit. Membrane fouling which results from deposition and nonspecific adsorption of proteins on the membrane surface is irreversible in nature, and results in a significant decrease in the membrane performance. To address this problem, two approaches were explored: 1) control of the surface chemistry tethering alumina membranes with organic components and 2) development of a novel photocatalytic membrane that exhibits hydrophilicity and can be easily regenerated. Both approaches can offer a viable route to the synthesis of attractive membranes, in that 1) the density of protein-resistant organic groups such as PEG is controllable by changing scaffolds or synthesis conditions and 2) the photocatalytic nanocomposite membranes can open the way for a new regeneration method that is environmentally benign.

Nanoporous materials have attracted tremendous interest, investment and effort in research and development due to their potential applications in various areas such as membranes, catalysis, sensors, delivery, and micro devices. Controlling a nanoporous material’s microstructure is of great interest due to the strong influence on efficiency and performance. For particles, microstructure refers to particle size, shape, surface morphology, and composition. When discussing thin films, microstructure includes film thickness, crystal orientation and grain boundaries. In this respect, we focus to develop novel methods for the synthesis and characterization of nanoporous materials and their films, which are capable of controlling the microstructure of material. This dissertation is composed of two main sections and each explores the fabrication of a different nanoporous material: 1) A simple fabrication method for producing oriented MFI zeolite membranes with controlled thickness, orientation, and grain boundary; 2) A microfluidic synthesis of ordered mesoporous silica particles with controllable size, shape, surface morphology, and composition. The first section of this dissertation demonstrates a simple and commercially viable method termed the micro-tiles-and-mortar method to make continuous b-oriented MFI membranes with controlled membrane microstructure. This simple method allows for control of the thickness of the membrane by using plate-like seed crystals with different thicknesses along the b-axis (0.5 μm to 2.0 μm), as well as to manipulate the density and structure of grain boundaries. Microstructural effects of silicalite-1 membranes on the gas separation are investigated by measuring the permeation and separation for xylene isomers. In the second section of this dissertation, a new synthesis method for the ordered mesoporous silica particles with controllable microstructure is demonstrated. This novel method combines a microfluidic emulsification technique and nonaqueous inorganic synthesis with a diffusion-induced self-assembly (DISA). The systematic control of the particle microstructure such as size, shape, and surface morphology is shown by adjusting microfluidic conditions.

none

Network-on-Chip (NoC) is communication infrastructure for future multi-core Systems-on-Chip (SoCs). NoCs are expected to overcome scalability and performance limitations of Point-to-Point (P2P) and bus-based communication systems. The routing algorithm of a given NoC affects the performance of the system measured with respect to metrics such as latency, throughput and load distribution. In this thesis, the popular Orthogonal One Turn (O1TURN) and Dimension Order Routing algorithms (DOR) for 2D-meshes are implemented by computer simulation. Investigating the effect of parameters such as packet, buffer and topology sizes on the performance of the network, it is observed that the center of the network is loaded more than the edges. A new routing algorithm is proposed and evaluated to achieve a more balanced load distribution. The results show that this goal is achieved with a trade off in latency and throughput in DOR and O1TURN.

In 2001, Fishburn, Tanenbaum, and Trenk published a pair of papers that introduced the notions of linear and weak discrepancy of a partially ordered set or poset. Linear discrepancy for a poset is the least k such that for any ordering of the points in the poset there is a pair of incomparable points at least distance k away in the ordering. Weak discrepancy is similar to linear discrepancy except that the distance is observed over weak labelings (i.e. two points can have the same label if they are incomparable, but order is still preserved). My thesis gives a variety of results pertaining to these properties and other forms of discrepancy in posets. The first chapter of my thesis partially answers a question of Fishburn, Tanenbaum, and Trenk that was to characterize those posets with linear discrepancy two. It makes the characterization for those posets with width two and references the paper where the full characterization is given. The second chapter introduces the notion of t-discrepancy which is similar to weak discrepancy except only the weak labelings with at most t copies of any label are considered. This chapter shows that determining a poset's t-discrepancy is NP-Complete. It also gives the t-discrepancy for the disjoint sum of chains and provides a polynomial time algorithm for determining t-discrepancy of semiorders. The third chapter presents another notion of discrepancy namely total discrepancy which minimizes the average distance between incomparable elements. This chapter proves that finding this value can be done in polynomial time unlike linear discrepancy and t-discrepancy. The final chapter answers another question of Fishburn, Tanenbaum, and Trenk that asked to characterize those posets that have equal linear and weak discrepancies. Though determining the answer of whether the weak discrepancy and linear discrepancy of a poset are equal is an NP-Complete problem, the set of minimal posets that have this property are given. At the end of the thesis I discuss two other open problems not mentioned in the previous chapters that relate to linear discrepancy. The first asks if there is a link between a poset's dimension and its linear discrepancy. The second refers to approximating linear discrepancy and possible ways to do it.

This dissertation has two principal components: the dimension of posets with planar cover graphs, and the cartesian product of posets whose cover graphs have hamiltonian cycles that parse into symmetric chains. Posets of height two can have arbitrarily large dimension. In 1981, Kelly provided an infinite sequence of planar posets that shows that the dimension of planar posets can also be arbitrarily large. However, the height of the posets in this sequence increases with the dimension. In 2009, Felsner, Li, and Trotter conjectured that for each integer h at least 2, there exists a least positive integer c(h) so that if P is a poset with a planar cover graph (the class of posets with planar cover graphs includes the class of planar posets) and the height of P is h, then the dimension of P is at most c(h). In the first principal component of this dissertation we prove this conjecture. We also give the best known lower bound for c(h), noting that this lower bound is far from the upper bound. In the second principal component, we consider posets with the Hamiltonian Cycle--Symmetric Chain Partition (HC-SCP) property. A poset of width w has this property if its cover graph has a hamiltonian cycle which parses into w symmetric chains. This definition is motivated by a proof of Sperner's theorem that uses symmetric chains, and was intended as a possible method of attack on the Middle Two Levels Conjecture. We show that the subset lattices have the HC-SCP property by showing that the class of posets with the strong HC-SCP property, a slight strengthening of the HC-SCP property, is closed under cartesian product with a two-element chain. Furthermore, we show that the cartesian product of any two posets from this strong class has the (weak) HC-SCP property.

Thesis (Ph. D.)--University of Missouri-Columbia, 1999. / Typescript. Vita. Includes bibliographical references (leaves 97-98). Also available on the Internet.