Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 2005. / Includes bibliographical references. / Chemical vapor deposition (CVD) produced films for a wide array of applications from a variety of organosilicon and organic precursors. The structure and properties of thin films were controlled by varying processing conditions such as the method and power of precursor activation, pressure, flow rates, and substrate temperature. Systematic variance of deposition conditions allows for the design of materials for a specific application, highlighting the versatility of CVD processes. Spectroscopic tools including Fourier transform infrared spectroscopy, variable angle spectroscopic ellipsometry, X-ray photoelectron spectroscopy, Raman spectroscopy, and nuclear magnetic resonance (NMR) spectroscopy were utilized to characterize film structure and understand the relationship between the structure and properties of materials. Computational quantum mechanics is a power tool applied to explain observed phenomena such as unreferenced chemical shifts in the 29Si NMR of organosilicon thin films, and to examine the thermochemistry of a family of methyl- and methoxymethylsilanes enabling the prediction of initial reactions occurring in the CVD process. / by Thomas Bryan Casserly. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 2006. / Includes bibliographical references (p. [189]-195). / Microfluidic devices and magnetorheological (MR) fluids have been two areas of intense research for several years. Traditionally, these two fields have remained separated from one another by scale. MR fluids are best known for their use in large-scale applications such as seismic dampers for the foundations of buildings. Recently however, there has been strong interest in utilizing the micro-structure formed by self-assembled MR fluids as structural components in microfluidic devices. MR fluids are composed of micron-sized paramagnetic colloids suspended in a non-magnetic carrier fluid. When an external magnetic field is applied to the MR fluid it induces dipole moments in the paramagnetic colloids causing them to interact and self-assemble into solid-like structures. In the past few years, self-assembled MR fluids have found use as structural components in a variety of micro-scale applications since their use provides an attractive alternative to traditional lithography as a means to create micro-scale structures. The advantages of using MR fluids include the ability to easily and inexpensively create sub-micron structures as well as the ability to create "dynamic structures" that can be controlled with an external magnetic field. / (cont.) In order for self-assembled MR fluids to be useful as structural components in microfluidic devices, we must understand the mechanisms and parameters governing their self-assembly under the confinement induced by the microfluidic channels. In this thesis we present a series of studies merging the areas of microfluidics and MR fluids. The systems we studied are technologically relevant as well as ideal for understanding the fundamental processes at work in the self-assembly of MR fluids under confinement. Using the Brownian dynamics simulation technique as well as experimental studies with microfluidic channels, we have illuminated many of the governing physical mechanisms involved in the self-assembly of MR fluids confined in microfluidic channels. In particular, we determined that the channel dimensions play a crucial role in dictating both the final structure as well as the dynamics of the MR fluid structures in these systems. We performed a systematic study of the structure and dynamics of MR fluids confined in two-dimensional (2D) channels with an external magnetic field directed normal to the channel in order to produce a state-diagram for the system as a function of the channel width and the magnetic field strength (the two controlling thermodynamic parameters). / (cont.) We further investigated the effects of the channel height and characterized the behavior of the system in the transition from quasi-2D to 3D microchannels. The behavior of the MR fluid structures in this transition region was found to depend heavily upon the volume fraction of the MR fluid and the height of the microchannel. We fully characterized the pore-size between clusters for dilute MR fluids self-assembled in quasi-2D microchannels, as it is an important parameter for microfluidic DNA mapping devices. Both the 2D and quasi-2D model systems provided valuable insight into the important mechanisms governing confined self-assembly. The impact of this work will be two-fold. Our study of the model systems of 2D channels and the transition from quasi-2D to 3D have made an important contribution to the literature on the physics of condensed matter systems. In particular, 2D colloidal systems are of great interest as models for self-assembly and phase transitions. Our work has illustrated how the presence of confining boundaries affects the behavior of 2D colloidal systems. Furthermore, the study of the quasi-2D system has shown how the interaction energy between MR fluid clusters dictates the structure that forms in this geometry. / (cont.) In addition to the fundamental scientific questions we have addressed, our work will also have a broad impact on the design of microfluidic devices utilizing self-assembled MR fluids. We have illustrated this design process with two examples but the possibilities for merging microfluidic device technology with MR fluids will provide many more applications in the years to come. / by Ramin Haghgooie. / Ph.D.

Thesis (Sc.D.) Massachusetts Institute of Technology. Dept. of Chemical Engineering, 1950. / Vita. / Bibliography: leaves 372-374. / by Edward Archibald Mason. / Sc.D.

Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 1929. / Includes bibliographical references (leaf 18). / by Milton H. Clapp, Ogden FitzSimons. / M.S.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 1995. / Includes bibliographical references (leaves 223-229). / by Erin M. Johnson. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 2001. / Includes bibliographical references. / The architecture of side-chain liquid crystalline diblock copolymers (SCLCBC's) involves order on two different but interdependent length scales. Through molecular-scale interactions, liquid crystalline moieties self-assemble into a variety of mesophases, and on a larger length scale, chemical differences between the two blocks can result in phase-segregated block copolymer microstructures. This combination of organizational tendencies offers experimentally and theoretically challenging problems that precede a wide array of applications. Using sequential anionic polymerization we synthesized a large number of well-defined SCLCBC's. These materials consist of an amorphous, polystyrene (PS) block and a methacrylate-based, LC block with side-chain mesogens that exhibit the Sc* mesophase. The samples' block copolymer morphology was evaluated using a combination of small angle X-ray scattering (SAXS) and electron microscopy, and thermal transitions were identified using polarized microscopy, calorimetry, and elevated temperature-SAXS. We found the formation of micron-sized domains and focal-conic superstructures to depend on the length of the PS block and the overall molecular weight. A morphological phase diagram was constructed based on LC volume fraction for molecular weights up to 40,000 Daltons. A broad lamellar regime was identified that extends to unusually low LC compositions. At intermediate LC fractions, just over 50 %-wt. LC, where one would expect lamellar morphologies for analogous coil-b-coil diblocks, morphologies are predominately lamellar, but exhibit perforated defects that connect the lamellar LC microdomains. / (cont.) A morphology consisting of hexagonally-packed PS cylinders was observed at 79 %-wt. LC, and at very high LC volume fractions (85 %-wt.) an unusual layered morphology was observed in both bulk and thin film studies. For this sample, the microstructure has a periodicity of -70 Angstroms and forms highly ordered micron-size monodomains. SAXS and TEM data suggest that the LC domains consist of smectic bilayers, and the amorphous polystyrene domains are highly oblate spheres that arrange hexagonally between smectic bilayers. We investigated the interdependence of block copolymer morphology and LC thermotropic phase behavior using elevated temperature SAXS. Order-disorder and order-order block copolymer transitions were located and compared to the LC isotropization temperature. For materials with low LC volume fractions, and low overall molecular weight, LC isotropization is shown to trigger morphological order-disorder transitions (ODT's). In other samples the LC clearing point precedes the ODT temperature, and this temperature difference largely depends on the length of the LC block. One sample, with 58 wt-% LC, undergoes a thermal-reversible LC triggered order-order transition between a predominately lamellar morphology with cylindrical defects and a completely lamellar morphology. These and other observations are explained on the basis of conformational asymmetry. Our analysis indicates that the length of the LC block and the related block copolymer superstructure are key parameters to controlling both LC mesophase and morphology. A simple free energy model was developed to capture the interplay between liquid crystalline and block copolymer morphology .... / by Mitchell Lewis Anthamatten. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 2011. / Cataloged from PDF version of thesis. / Includes bibliographical references. / Over the past few decades, increase in the demand for low molecular weight alcohols, like ethanol and butanol, for use as a biofuel has provided a new impetus to their production by the fermentation of polysaccharides, and the subsequent separation from the alcohols from the aqueous fermentation broth. The inhibitory nature of alcohols to their own production necessitates the continuous lowering of the concentration of the alcohol during the fermentation process. The technology for removing alcohol and other organics from aqueous solutions also finds application in industrial waste treatment facilities. The different techniques in use currently for in-situ removal of alcohol from fermentation broth, like distillation, suffer from drawbacks like high energy consumption. The goal of this project was to develop a redox-responsive polymer which has preferential selectivity for butanol causing the polymer to selectively extract butanol from aqueous fermentation broth. On application of electric potential, the redox moieties in the polymer were oxidized and charged resulting in an increase in the hydrophilicity of the chemical environment in the gel and the extracted butanol was released into a stripping medium. The switchable selectivity of the polymer for butanol allows its use for the development of continuous separation system for butanol extraction. In this project, novel co-polymer of hydroxybutyl methacrylate (HBMA) and vinylferrocene (VF) was synthesized by free radical polymerization. The HBMA backbone gave the polymer preferential selectivity for butanol, while the ferrocene (Fc) groups of VF made the polymer redox active. The thermodynamic parameters, equilibrium distribution coefficient and separation factor, which quantify the distribution of a species between two phases were experimentally determined for butanol and water distribution between the polymer and the aqueous phase when the redox moieties in the polymer were in the reduced and the oxidized states respectively. The values of these parameters confirmed that the oxidation and the consequent charging of the redox species resulted in a decrease in the polymer's affinity for hydrophobic molecule, butanol. The optimum composition of the co-polymer was arrived at by comparison of properties of polymers with different compositions. The redox active co-polymer of HBMA and VF was attached to electrically conducting substrates to prepare redox polymer electrodes (RPEs). The RPEs allowed the oxidation and reduction of the ferrocene groups in the polymer by the application of electric potential. Carbon black (CB) and carbon fiber mats, called carbon paper (CP) were used as the substrates. RPEs were prepared using five different techniques-three techniques were based on strategies reported in literature and involved the chemical modification of the functional groups on the surface of CB and CP to allow polymer grafting. In addition, an iCVD based technique was developed which functionalized the CP surface by deposition of a reactive polymer, poly(pentafluorophenyl methacrylate (PFM)-co-ethyleneglycol diacrylate (EGDA)). The polymer layer was chemically modified to allow redox polymer grafting.. Impregnation of porous CP with redox polymer gel resulted in electrodes with highest mass of polymer per unit mass of conducting substrate. / (cont.) The electrochemical activity and reversibility of the RPEs prepared using the different techniques were ascertained by cyclic voltammetry. The impregnated electrodes were used to demonstrate the successful use of the polymer gel to extract butanol from its aqueous solution, and release it into water upon oxidation. Conceptual scheme of a continuous separation system that can be built using these RPEs was proposed and the separation that can be achieved using such a system was determined by simulating the continuous separation process using finite element modeling. It was determined that the separation system integrated with a fermentation reactor can help maintain the concentration of butanol at a value 20% lower than the critical value beyond which fermentation is completely inhibited. The mechanism of electron transport in the polymer coated RPEs was investigated. Diffusion of electrons was found to be the rate controlling step. Further, it was found that diffusion of electrons due to the 'hopping' of electrons from one redox site to the next, and the electronic motion due to the motion of the polymer chains themselves played important roles in determining the apparent diffusivity of electrons. As part of the PhDCEP Capstone project, the potential of butanol produced through fermentation, commonly known as bio-butanol, was analyzed as a blend for gasoline was analyzed. It was found that although the market for gasoline blend is huge and growing, butanol suffers from higher cost of production with respect to its primary competitor, bio-ethanol. Chances of bio-butanol's potential success can be enhanced through a combination of technological breakthroughs including development of strains of high yield bacteria, use of inexpensive lignocellulosic biomass, and process design improvements. / by Abhinav Akhoury. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 2013. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 114-117). / Electrospinning is a convenient method to produce nanofibers with controlled diameters on the order of tens to hundreds of nanometers. The resulting nonwoven fiber mats are lightweight, highly porous, and have high specific surface areas around 1 to 100 m2/g. Combined with the high electrical conductivity of intrinsically conductive polymers, conductive electrospun fiber mats are promising for a variety of applications, such as multifunctional textiles, resistance-based sensors, flexible reversibly hydrophobic surfaces, organic photovoltaics, scaffolds for tissue engineering, and conductive substrates for surface functionalization and modification Intrinsically conductive polymers, such as polyaniline (PAni), however, are relatively hard to Intrinsically conductive polymers, such as polyaniline (PAni), however, are relatively hard to process compared to most other polymers. They have fairly rigid backbones due to the high aromaticity, and are usually available only in relatively low molecular weight forms, so that the elasticity of their solutions is insufficient for it to be electrospun directly into fibers. Considerable amount of recent work has been reported trying to make electrospun polymeric nanofibers with intrinsically conductive polymers or composites. However, a large fraction of the work only showed the morphology and did not characterize the actual performance of these fibers, nor did they test the variability of the fibers and mats from a wide range of processing conditions and resulting structures. Therefore, this thesis aims to make a comprehensive study of the electrical tunability of electrospun fibers with intrinsically conductive polymers and its composites, to establish a clear processing-structure-property relationship for these fibers and fiber mats, and to test the resultant fibers with the targeted applications such as gas sensing. We have first developed a reliable method to characterize fiber electrical conductivity using interdigitated electrodes (IDE) and high-impedance analyzers with contact-resistance corrections, and applied to electrospun conductive polymer nanofibers. This method was shown to be reliable and sensitive, as opposed to some of the other methods that have been reported in literature. Facing with the challenge of overcoming the relatively low elasticity of the conductive polymer solutions to achieve electrospinnability, we have fabricated electrospun fibers of PAni and poly(3,4-ethylenedioxythiophene) (PEDOT), blended with poly(ethylene oxide) (PEO) or poly(methyl methacrylate) (PMMA) over a range of compositions. Pure PAni (doped with (+)- camphor-i 0-sulfonic acid (HCSA)) fibers were successfully fabricated for the first time by co-axial electrospinning and subsequent removal of the PMMA shell by dissolution. This allowed for the pure electrospun PAni/HCSA fibers to be tested for electrical performances and its enhancement as well as gas sensing application. The conductivities of the PAni-blend fibers are found to increase exponentially with the weight percent of doped PAni in the fibers, to as high as 50 ± 30 S/cm for as-electrospun fibers of 100% PAni/HCSA. This fiber conductivity of the pure doped PAni fibers was found to increase to 130 ± 40 S/cm with increasing molecular orientation, achieved through solid state drawing. The experimental results thus support the idea that enhanced molecular alignment within electrospun fibers, both during the electrospinning process and subsequent post-treatment, contributes positively to increasing electrical conductivity of conductive polymers. Using a model that accounts for the effects of intrinsic fiber conductivity (including both composition and molecular orientation), mat porosity, and the fiber orientation distribution within the mat, calculated mat conductivities are obtained in quantitative agreement with the mat conductivities measured experimentally. This correlation, along with the reliable method of fiber conductivity measurement by IDE, presents a way to resolve some of the inconsistencies in the literature about reporting electrical conductivity values of electrospun fibers and fiber mats. Pure PAni fibers with different levels of doping were also fabricated by co-axial electrospinning and subsequent removal of the shell by dissolution, and shown to exhibit a large range of fiber electrical conductivities, increasing exponentially with increasing ratio of dopant to PAni. These fibers are found to be very effective nanoscale chemiresistive sensors for both ammonia and nitrogen dioxide gases, thanks to this large range of available electrical conductivities. Both sensitivity and response times are shown to be excellent, with response ratios up to 58 for doped PAni sensing of ammonia and up to more than 105 for nitrogen dioxide sensing by undoped PAni fibers. The characteristic times for the gas sensing are shown to be on the order of 1 to 2 minutes. We have also developed a generic time-dependent reaction-diffusion model that accounts for reaction kinetics, reaction equilibrium, and diffusivity parameters, and show that the model can be used to extract parameters from experimental results and used to predict and optimize the gas sensing of fibers under different constraints without the need to repeat experiments under different fiber and gas conditions. / by Yuxi Zhang. / Ph.D.

Thesis: Sc. D., Massachusetts Institute of Technology, Department of Chemical Engineering, 2014. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 131-151). / Over the past 30 years, advances in nanotechnology have generated a multitude of nanostructures exhibiting a breadth of physical, chemical, and biological properties that have tremendous potential to improve the detection and treatment of disease. Despite this progress, biomedical nanotechnologies have yet to approach the same level of complexity as biological systems, which produce higher-order functions through coordinated interactions between multiple nanoscale components. This thesis aims to explore the potential of nanoparticles to interface with the host biology to perform systems-level applications that benefit disease sensing and treatment. First, we engineered nanoparticles to sense dysregulated protease activity associated with thrombosis and generate reporters that can be noninvasively quantified in the urine. These nanoparticles exploit the vascular transport of the circulatory system and the size filtration function of the renal system to emit reporters into the urine following proteolytic cleavage events. The reporter levels in the urine differentiate between healthy and thrombotic states and correlate with clot burden in a mouse model of pulmonary embolism. Next, we developed nanoparticles that homeostatically regulate the biological cascade responsible for haemostasis to prevent the aberrant formation of clots. These nanoparticles form a negative feedback loop with thrombin, a key enzyme in the coagulation cascade, to regulate their release of the anticoagulant heparin. In mice, they inhibited the formation of pulmonary embolisms without an associated increase in bleeding, the primary side-effect of antithrombotic therapy in the clinic. Finally, we investigated a two-component system whereby the first therapeutic entity induces the upregulation a molecular signal within a malignant environment to amplify the local recruitment of a secondary population of targeted nanoparticles. Here, the interaction between the initial therapeutic and the targeted nanoparticles occurred indirectly through a biological stress pathway. This cooperative targeting system delivered up to five-fold higher nanoparticle doses to tumors than non-cooperative controls, leading to delayed tumor growth and improved survival in mice. Together, these systems highlight the potential for interactive nanoparticle systems to perform highly complex functions in vivo by leveraging and modulating the host biology. In contrast to the current strategy of injecting large populations of nanoparticles that carry out identical, pre-defined tasks with little to no feedback from the in vivo environment, this work supports the construction of nanoparticle systems that leverage both synthetic and endogenous components to produce emergent behaviors for enhancing diagnostics and therapeutics. / by Kevin Lin. / Sc. D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 2007. / Includes bibliographical references. / In pharmaceutical manufacturing practices, blending of active pharmaceutical ingredient (API) with excipients is a crucial step in that homogeneity of active ingredient after blending is a key issue for the quality assurance of final products. Inadequate knowledge of the interdependence between raw material properties and their impact on the blending process often gives rise to product variance and failure and therefore higher manufacturing costs. Since particles are the basic unit of powder flow, a fundamental understanding of the crucial particle characteristics and particle interactions is essential for a good prediction and control of the blending process. In this work, inter-particle adhesion forces including van der Waals force, capillary force and electrostatic force of lactose monohydrate and microcrystalline cellulose were measured by the atomic force microcopy and other techniques. Their correlations with particle properties and environmental variables were elucidated quantitatively through mathematical modeling, and their impacts on powder blending homogeneity were investigated experimentally. It was found that surface roughness, electrostatic surface charges, moisture sensitivity as well as relative humidity are crucial parameters to determine inter-particle adhesion forces. / (cont.) By controlling these factors, the inter-particle adhesion forces can be optimized to improve final blend homogeneity. For instance, using excipient particles processed with surface-smoothing method reduced the blending time to reach endpoint. It was also found that enhancing electrostatic attractive interactions between excipient and API particles resulted in better blend homogeneity. In addition, the mathematical force models developed in this study allowed us to predict the magnitudes of inter-particle adhesion forces, which can be later used as an important input parameter in simulating the powder blending process of different scales. The mechanistic knowledge of particle interactions and their dependence on particle properties through this study provides a theoretical foundation for a successful linkage between the micro-scale particle level and the macro-scale bulk powder flow behavior, enhances process understanding, and opens opportunities for process improvement. / by Yu Pu. / Ph.D.