Thesis: Ph. D., Massachusetts Institute of Technology, Department of Chemical Engineering, 2017. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 153-167). / Messenger RNA (mRNA) therapeutics have the potential to treat a diverse array of diseases requiring protein expression, with applications in protein replacement therapies, immunotherapies, and genome engineering. However, the intracellular delivery of mRNA is challenging and necessitates a safe and effective delivery vector. Lipid nanoparticles (LNPs) have shown considerable promise for the delivery of small interfering RNAs (siRNA) to the liver but their utility as agents for mRNA delivery have only been recently investigated. New delivery materials for mRNA delivery are also being developed which have the potential to transfect nonliver targets, but the screening of these vectors in vivo is low-throughput and it is difficult to determine transfected cell types. There is a need both for efficacious, well-characterized mRNA delivery materials and for methods to facilitate in vivo screening of novel materials. We first developed a generalized strategy to optimize LNP formulations for mRNA delivery to the liver using Design of Experiment methodologies. By simultaneously varying lipid ratios and structures, we developed an optimized formulation which increased the potency of eryrthopoietin-mRNA-loaded LNPs in vivo 7-fold relative to formulations previously used for siRNA delivery. Next, we explored the immune response and activity of base-modified LNPformulated mRNA administered systemically in vivo. We observed indications of a previously uncharacterized transient, extracellular innate immune response to mRNA-LNPs, including neutrophilia, myeloid cell activation, and up-regulation of four serum cytokines. Although we have developed a more efficacious liver-targeting LNP, many mRNA therapies will require delivery to non-liver tissues. Using trial-and-error approaches, we discover novel formulations capable of inducing mRNA expression in vivo in the spleen, lung, and fat. To increase the throughput of in vivo screening, we report a new barcoding-based approach capable of evaluating the biodistribution and pharmacokinetics of many LNP formulations in a single mouse. Then, we develop a method that can identify mRNA expression delivered from LNPs in both bulk tissues and with single cell resolution. Together, the work reported here contributes to the development of mRNA therapeutics by increasing mRNA-LNP potency and characterizing their immunogenicity in vivo. Furthermore, we hope the multiple in vivo screening methods described in this Thesis will accelerate the discovery of new delivery vectors capable of transfecting desired tissues and cell types. / by Kevin John Kauffman. / Ph. D.

Thesis (Sc. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 1983. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND SCIENCE / Bibliography: leaves 341-347. / by Shan Hsi Yang. / Sc.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 2009. / Includes bibliographical references. / Liquid Crystalline Elastomer (LCE)-inspired segmented polyurethane elastomers possessing widely different extents of ordering were created to mimic the hierarchical structure of the continuous matrix and superior mechanical performance of spider silk fibers. The silk's remarkable toughness originates from a fiber morphology that possesses [beta]-pleated crystalline sheets within an amorphous matrix. In the polyurethane materials, various extents of poly(ethylene oxide) (PEO) soft segment ordering were implemented within continuous soft domains that were connected by hexamethylene diisocyanate-butanediol (HDI-BDO) hard segments. Soft segment crystallinity studies revealed the need to optimize the extent of continuous soft domain ordering. Highly crystalline PEO soft segments, while they display good microphase segregation properties, sacrifice extensibility due to their high melting transition temperature. Moderately crystalline PEO soft segments, meanwhile, possess less defined phase segregation but enhanced mechanical properties from their reversible dispersed crystalline soft segment domains. Non-crystalline Pluronic copolymer systems had good mechanical properties that resulted from both a strong hard segment incompatibility and a highly mobile soft segment matrix. Hydrogen-bonded hard domain shearing during in-situ tensile deformation yields oriented hard blocks that align at a preferred tilt angle of ±60° from the strain direction. Extensive alignment and orientation of the moderately-ordered PEO soft segments occurred during deformation, which was consistent with its observed mechanical behavior. Pluronic-containing segmented polyurethanes formed an ordered mesophase in the continuous soft matrix during deformation. A series of cyclic, aliphatic polyurethanes with dicyclohexyl methane diisocyanate (HMDI) hard segments and poly(tetramethylene oxide) (PTMO) soft segments was synthesized to study compositional effects on the extent of soft segment mixing, and how these effects translated to both mechanical and barrier performance. Shorter soft segment chain systems displayed a greater hard segment compatibility, which resulted in materials that were both more rigid mechanically and provided better barrier characteristics. / (cont.) Longer soft segments in the continuous polymer matrix displayed a more phase segregated structure, which enhanced their mechanical properties but sacrificed barrier effectiveness. Incorporation of dimethyl propane diol (DMPD), a branched chain extender, created a completely amorphous polyurethane matrix. Polyurethane/Laponite nanocomposites were also created using particles that were capable of preferentially associating with hard or soft segments. HMDI-BDO-PTMO polyurethane/Laponite nanocomposites demonstrated drastically reduced mechanical performance (~13-fold decrease in toughness and ~10-fold decrease in extensibility). The deteriorated mechanical performance was attributed to the formation of an interconnected hard segment continuous morphology that significantly reduced matrix extensibility. HMDI-DMPDPTMO polyurethane/Laponite composites, on the other hand, only experienced modest reductions in extensibility (-70% of total initial extensibility) while maintaining toughess and increasing initial modulus 10-fold. Mechanical behavior resulted from well-dispersed Laponite clay platelets that reinforced the amorphous polymer matrix while imposing modest chain segmental mobility restrictions. / by Ryan Scott Waletzko. / Ph.D.

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Chemical Engineering, 2016. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 189-198). / Indium phosphide (InP) quantum dots (QDs) stand out as the most promising candidate to replace the currently commercialized cadmium-containing materials for optoelectronic applications. This thesis focuses on using experimental and theoretical methods to study growth mechanisms of InP QDs from precursor conversion to final nanocrystal formation. As the key experimental platform, a high temperature and high pressure microfluidic system was first applied to study the effect of group V precursor reactivity on the QD growth. High-pressure flow conditions allow for precise control of synthetic parameters and also the use of low-boiling-point solvents for synthesis with enhanced mixing. Results showed that lowering the precursor reactivity did not significantly improve the QD quality, contradicting the original hypothesis. The unexpected role of precursor chemistry motivated investigation into the early-stage QD growth mechanisms. First-principles approaches were used without any prior assumptions on reaction pathways. Simulations showed that small clusters with indium-rich surfaces form in the early-stage QD growth. In and P precursors have different roles, with P precursors controlling the reaction energy, and In precursors determining the reaction barrier. With clusters identified as important growth intermediates in both simulations and experiments, their role during the QD formation was then investigated with a one-solvent protocol, which combined flow synthesis, GPC purification and MALDI mass characterization. Experiments revealed that similar clusters exist during the late-stage nanocrystal growth, suggesting their role as a continuous supply for the QD formation. Lastly, a QD size tuning strategy was developed involving the use of weakly associated ligands to synthesize cluster-free InP QDs with different sizes and narrow size distributions. This synthetic approach enabled the construction of a correlation between the absorption features and the mass and concentration of InP QDs. The importance of In precursor quality became apparent after exploring effects of impurities and solvents. For example, when water and hydroxide/oxide species contaminate In precursors, the growth of InP QDs are inhibited and batch-to-batch variations are observed. / by Lisi Xie. / Ph. D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 1993. / Includes bibliographical references (p. 214-219). / by Thomas Andrew Meadowcroft. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 2009. / "May 15th, 2009." / Includes bibliographical references (leaves 246-249). / Differential equations are one of the primary tools for modeling phenomena in chemical engineering. While solution methods for many of these types of problems are well-established, there is growing class of problems that lack standard solution methods: partial integro-differential equations. The primary challenges in solving these problems are due to several factors, such as large range of variables, non-local phenomena, multi-dimensionality, and physical constraints. All of these issues ultimately determine the accuracy and solution time for a given problem. Typical solution techniques are designed to handle every system using the same methods. And often the physical constraints of the problem are not addressed until after the solution is completed if at all. In the worst case this can lead to some problems being over-simplified and results that provide little physical insight. The general concept of exploiting solution domain knowledge can address these issues. Positivity and mass-conservation of certain quantities are two conditions that are difficult to achieve in standard numerical solution methods. However, careful design of the discretizations can achieve these properties with a negligible performance penalty. Another important consideration is the stability domain. The eigenvalues of the discretized problem put restrictions on the size of the time step. For "stiff' systems implicit methods are generally used but the necessary matrix inversions are costly, especially for equations with integral components. By better characterizing the system it is possible to use more efficient explicit methods. / (cont.) This work improves upon and combines several methods to develop more efficient methods. There are a vast number of systems that be solved using the methods developed in this work. The examples considered include population balances, neural models, radiative heat transfer models, among others. For the capstone portion, financial option pricing models using "jump-diffusion" motion are considered. Overall, gains in accuracy and efficiency were demonstrated across many conditions. / by Alexander M. Lewis. / Ph.D.

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Chemical Engineering, February 2017. / "September 28th, 2016." Cataloged from PDF version of thesis. / Includes bibliographical references (pages 127-133). / Semiconductor nanocrystals, also known as quantum dots, are an exciting class of materials because their band gap can be tuned according to the nanocrystal size. In this way, the material band gap can be largely decoupled from its atomic composition - a property unique to this system. The potential applications for semiconductor nanocrystals are wide ranging and include: LEDs, photovoltaics, photon downconversion, photon upconversion, and thermoelectrics. However, their size-dependent band gap can also be a hindrance, as any size variation in the ensemble of nanocrystals introduces energetic disorder and spatial disorder in films. While synthesized as a colloid, for most applications the nanocrystals are deposited as a thin film. The rate of energy transfer between nanocrystals in the film, dictated by the arrangement and distance between neighbors, is therefore a critical parameter affecting device efficiency. As a result, controlling the nanocrystal physical arrangement is crucial to the success of these materials. Despite this, there is a lack of understanding of how to observe and control these processes at the nanoscale. This thesis begins by improving the synthesis of lead sulfide (PbS) nanocrystals to produce narrow size dispersity ensembles with tunable average size by ensuring the reaction is diffusion-limited. We then experimentally determine what parameters (ligand coverage, solvent, size dispersity) most affect the ability of these nanocrystals to self-assemble into highly ordered superlattice structures. We show that superlattices can be produced with a wide variety of surface ligands of differing lengths, either directly from a colloidal suspension or post-deposition and we thoroughly characterize the interparticle spacing as a function of ligand species. Next, we demonstrate an in situ X-ray scattering technique which enables the real-time visualization of nanocrystal self-assembly, with details unprecedented by any other experimental method. This technique led to a better understanding of the colloid to superlattice transition, including the observation of intermediate states and the ability to compare kinetics of different self-assembly aspects. Finally, we present experimental measurements demonstrating that nanocrystal size dispersity and selfassembly are critical to efficient energy transfer in films and that as energetic disorder is minimized through improved synthetic methods, spatial disorder becomes an increasingly important parameter to control. In the final experimental chapter of this thesis, we apply this knowledge to a different material system perovskite nanoplatelets, which have the potential to be useful as an inexpensive, solution-processable emission material. For these 2D materials, we optimize the thickness homogeneity and study the selfassembly of the nanoplatelets into stacked superstructures. We highlight the incredible tunability of this material system accessible through thickness and compositional tuning, which allows absorption and emission to be shifted across the entire visible range. Lastly, we demonstrate the potential of this system for next generation LEDs. / by Mark Clayton Weidman. / Ph. D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 1995. / Includes bibliographical references. / by Hiroshi Harlan Saito. / Ph.D.

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Chemical Engineering, 2017. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 155-162). / The production of phosphoric acid from phosphate mineral rock involves the addition of phosphate rock to a concentrated sulfuric acid solution. The induced reactive crystallization process produces a side product of calcium sulfate hydrates, which become the filter media in the subsequent acid separation process. For most industrial processes, the dihydrate form of calcium sulfate crystals (gypsum) precipitates and its shape and size distribution are key factors in determining the downstream filtration efficiency. Particularly, the metal ion impurities coming from raw phosphate rock play an important role as shape modifiers. The presence of impurities in the acid mixture has an impact both thermodynamically and kinetically, although most of the available literature focuses on their sole role as growth inhibitors and has neglected their potential impact on altering solution speciation. Past studies on gypsum crystallization in phosphoric acid solutions usually involve the study of crystal growth and nucleation kinetics. However, most of these works did not use the correct definition of supersaturation when quantifying kinetic parameters. The high concentrations in this multicomponent electrolyte system implies that supersaturation, which be written in terms of the solubility product ratio, as governed by nonideal thermodynamics, requires the computation of activity coefficients as well as free ion concentrations. For this purpose, the mixed solvent electrolyte (MSE) model is utilized to capture the solution speciation in order to properly quantify supersaturation at any given condition. The MSE model is a first-principles model that determines solid-liquid equilibrium by calculating excess Gibbs energy from additive pairwise interactions. When impurities are present, additional binary interactions need to be included in the databank, which is carried out by regression analysis using solubility measurements. Continuous reactive crystallization experiments are carried out with and without additives using a mixed-suspension, mixed-product removal (MSMPR) crystallizer. The crystal size distribution and supersaturation are measured once the process reaches steady state. Different conditions are imposed to acquire both the temperature and supersaturation dependency of the crystallization kinetics. A two-dimensional growth model with dispersion is developed in order to capture the needle-like crystal morphology and the temperature dependence of the crystal aspect ratio, which is made possible by performing multi-scale image segmentation and edge detection using the Canny method. Experimental and numerical results are obtained for the base system and in the presence of single and combined impurity ions. Different growth inhibition models are verified and compared for numerical quantification of step advancement retardation in the presence of impurities. This study goes beyond past studies by providing a full two-dimensional kinetic model for a highly concentrated ionic system that includes crystallization kinetics and a thermodynamically correct driving force accounting for non-ideality as well as the effects of impurities. / by You Peng. / Ph. D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 2004. / Page 178 blank. / Includes bibliographical references. / (cont.) the feed fluid was achieved after one pass through the counter current system. In the second case, four permanent magnets were arranged in a quadrupole around a central column to create areas of high magnetic field at the column walls and areas of low magnetic field at the centerline, inducing non-magnetic particles to concentrate at the centerline, where they were removed through a coaxial central outlet tube at the top of the column. Depending on the flow rate, up to 99% of polystyrene beads of different sizes could be removed from the feed after one pass through the quadrupole system. The recovery efficiency decreased with increasing flow rate, i.e. with decreasing residence time in the device. E. coli cells were able to be removed with separation efficiencies as high as 95% at much higher flow rates due to the formation of [approximately]12 micron aggregates in the presence of the magnetic nanoparticles; these large aggregates experienced enhanced magnetic forces over individually-dispersed cells and could be recovered more effectively. The governing equations for magnetophoretic clarification were applied to the quadrupole configuration to predict particle trajectories through the column and to predict the separation efficiency under different flow conditions, which showed a good match to the experimental results. It was also shown that axial magnetic field gradients near the entrance region acted effectively as a barrier to entry of particles in the slow moving regions near the walls; this retardation of their axial movement provided a longer residence time for the particles that allowed them to be moved more efficiently to the centerline ... / A new approach for the removal of micron-sized particles from aqueous suspensions was developed and applied to the problem of cell clarification from raw fermentation broth. The concepts of magnetophoretic separation were exploited to take advantage of the force that acts on a non-magnetic particle when it is immersed in a magnetic fluid (ferrofluid) that is subjected to a non-uniform magnetic field. The magnetic "pressure" difference across the non-magnetic particle owing to the magnetization of the surrounding magnetic fluid forces the particles away from areas of high magnetic field strength and into areas of low magnetic field strength. This force is proportional to the volume of the non-magnetic particles, and is therefore stronger for larger particles. In this way, non-magnetic particles can be focused and moved out of the bulk fluid by applying a non-uniform magnetic field to the system, leading to magnetophoretic clarification. The magnetic fluid used in this work was composed of magnetite nanoparticles coated with a poly(acrylic acid)-poly(ethylene oxide)-poly(propylene oxide) graft copolymer layer that stabilized the nanoparticles in water and prevented their aggregation. The magnetic nanoparticles were approximately 32 nm in diameter, with the magnetite core itself being approximately 8 nm in diameter. Magnetophoretic clarification was investigated using two different flow configurations. In the first case, the particle-laden magnetic fluid was pumped through a flow tube while a series of magnets around the tube moved counter to the direction of the feed flow; the non-magnetic particles in the feed were captured and effectively removed from the bulk fluid by the moving magnets. A removal efficiency of 95% of E. coli cells from / by Sonja Ann Sharpe. / Ph.D.