Thesis: Ph. D., Massachusetts Institute of Technology, Department of Materials Science and Engineering, 2016. / This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. / Cataloged from student-submitted PDF version of thesis. / Includes bibliographical references (pages 121-139). / In order for alkali-ion batteries to gain widespread adoption as the energy storage technology of choice for transportation and grid applications, their energy must be improved. One key step towards this necessary improvement is the development of new battery cathode materials. In this thesis, two classes of polyanionic materials are examined as candidate cathodes for alkali-ion batteries: Li-containing carbonophosphates for Li-ion batteries and Na-containing fluorophosphates for Na-ion batteries. High-throughput ab initio calculations have previously identified carbonophosphates as a new class of polyanionic cathode materials. Li₃MnCO₃PO₄ is the most promising candidate due to its high theoretical capacity, predicted multi-redox activity, and ideal voltage range. However, a major limitation of this material is its poor cyclability and experimental capacity. In this work Li₃Fe₀.₂Mn₀.₈CO₃PO₄ is synthesized to combine the high theoretical capacity of Li₃MnCO₃PO₄ with the high cyclability of Li₃FeCO₃PO₄. Li₃Fe₀.₂Mn₀.₈CO₃PO₄ outperforms both Li₃MnCO₃PO₄ and Li₃FeCO₃PO₄, showing a reversible capacity of 105 mAh/g with little capacity fade over 25 cycles. However, poor thermodynamic stability of these compounds, particularly at partially delithiated compositions, prevents carbonophosphates from being seriously considered as a viable Li-ion cathode. Fluorophosphate cathodes are currently one of the most promising polyanionic sodium-ion battery cathodes due to their high energy density and cyclability. To further improve fluorophosphate cathodes, their capacity must be increased by using Na sites that had not been accessed prior to this work. In this thesis, reversible electrochemical Na+ insertion into Na₃V₂(PO₄)₂F₃ is demonstrated. To further improve fluorophosphate cathodes by using its newly discovered insertion capacity, novel Na₃[M]₂(PO₄)₂F₃ cathodes, with {M = Fe, Ti, V}, are synthesized and evaluated. Seeing no improvement, the question of what specific mechanism limits fluorophosphate cathode capacity is addressed. For this, the synthesis, electrochemical characterization, and computational examination of a specifically designed test system, Na₃GaV(PO₄)₂F₃, is reported. This leads to the conclusion that large diffusion barriers at high sodiations impose a kinetic limit on Na+ insertion in fluorophosphate cathodes, as opposed to limits in transition metal redox activity. / by Ian Lawrence Matts. / Ph. D.

Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Materials Science and Engineering, 1997. / Includes bibliographical references (p. 66). / by Ingchie N. Kwan. / M.S.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Materials Science and Engineering, 2011. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 155-169). / Structural biological materials such as bone, nacre, insect cuticle, and sea sponge exoskeleton showcase the use of inferior building blocks like proteins and minerals to create structures that afford load-bearing and armor capabilities. Many of these are composite structures that possess the best of the properties of their base constituents. This is in contrast to many engineering materials, such as metals, alloys, ceramics and their composites which show improvement in one mechanical property (e.g. stiffness) at the cost of another disparate one (e.g. toughness). These excellent design examples from biology raise questions about whether similar design., and improvement in disparate properties, can be achieved using common engineering materials. The identification of broad design principles that can be transferred from biological materials to structural design, and the analysis of the utility of these principles have been missing in literature. In this thesis, we have firstly identified certain universal features of design of biological structures for mimicking with engineering materials: a) presence of geometric design at the nanoscale, b) the use of mechanically inferior building blocks, and c) the use of structural hierarchies from the nanoscale to the macroscale. We firstly design. in silico, metal-matrix nanocomposites, mimicking the geometric design found at the nanoscale in bone. We show this leads to improvements in flow strength of the material. A key finding is that limiting values of certain of these parameters shuts down dislocation-mediated plasticity leading to peak in flow strength of the structure. Metals are however, costly constituents, and we next confront the issue of whether it is possible to use a single mechanically inferior and commonly available constituent, such as silica, to create superior bioinspired structures. We turn to diatom exoskeletons, protective armor structures for algae made almost entirely of silica, and create nanoporous silica structures inspired from their geometry. We show large improvements in ductility of silica through this design, facilitated by a key size-dependent brittle-to-ductile deformation transition in these structures. Nanostructuring, while improving ductility, affects the stiffness of these structures, softening them by up to 90% of bulk silica. Hierarchical assembly of silica structures is then used to regain the stiffness lost due to nanostructuring while not losing their improvement in toughness. Finally, improvement in toughness with several levels of hierarchy is studied, to showcase a defect-tolerant behavior that arises with the addition of hierarchies, i.e., tolerance of the fracture strength to a wide range of sizes of cracks present in the structure. The importance of R-curve behavior, i.e., toughness change with the advance of a crack in the structure. to the defect-tolerance length scale is also established. These findings showcase the validity of using design principles obtained from biological materials for improvement in mechanical properties of engineering materials. / by Dipanjan Sen. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Materials Science and Engineering, 1988. / Vita. / Includes bibliographical references. / by Mark John Gore. / Ph.D.

Polymer-clay nanocomposites are attracting global interest principally because property enhancements are obtained at low clay particle loadings (1-5 wt%). However there is lack of fundamental understanding of such composites. The aim of this work is to provide an insight into the interaction between polymer and clay. This includes the driving force for intercalation, the reinforcement mechanisms and property-volume fraction relationships. Functionalised poly(ethylene glycol)-clay, poly(c-caprolactone)-clay and thermoplastic starch-clay nanocomposites with a range of polymer molecular weights, clay volume fractions and with different clays were prepared using solution methods, melt-processing methods, and in situ polymerisation. A reliable X-ray diffraction technique for low angle basal plane spacing of clay, the essential parameter for structure determination, was established obtaining ±0.005 Mn between three diffractometers. The basal plane spacing was found to be unaffected by polymer molecular weight and preparation method but was affected by the nature of the polymer and clay. Increasing clay loading could lead to a lower spacing. As a cautionary observation, poly(ethylene glycol) with high molecular weight (2: 10,000) was found to undergo degradation readily during preparation of nanocomposites with and without clay. Competitive sorption experiments for molecular weight showed that high molecular weight fractions of polymer intercalate preferentially into clay during solution preparation. Thermodynamic studies on the intercalation process found that significant enthalpic change occurred during intercalation, which is coincident with the observation that heat-treated clays without interlayer water can intercalate polymer. The calculation of true volume fraction against nominal volume fraction provided reasonable explanation of property enhancement and helps understand the relation between nanocomposites and conventional composites. At a given clay loading, nanocomposites with better dispersion gave more property enhancement than those with lower dispersion or conventional composites. The crystallinity of semicrystalline polymer was also affected by varying extents of dispersion of clay. The use of X-ray diffraction with an internal standard was explored for quantitative analysis of intercalation and exfoliation.

Adhesive bonded applications are used widely in industry because of significant advantages such as uniform stress distribution, and the ability to join different materials. However most epoxy structural adhesives are brittle at room temperature and it is required to improve their toughness. The objective of this work was to understand the fracture of adhesive joints, failure criteria and rubber toughening mechanisms via a series of experiments and FEA modelling. Double lap joints (DLJ) bonded by commercial AV119 adhesive were studied. It was found that local strain and failure path were controlled by adhesive thickness. In order to model adhesive joints accurately and efficiently, systematic fracture tests were implemented to determine the fracture criteria. Mode-I, mode-II and mixed mode fracture energy release rates were obtained by Fixed Arm Peel, 4-point End Notched Flexure (ENF) and Mixed Mode Bending (MMB) tests. Numerical analysis was applied to determine the parameters of the Drucker-Prager material model and Cohesive Zone Model (CZM). The 3D FEA results showed good agreement with experimental results of DLJ and MMB. FEA results successfully demonstrated bonding strength, stress and strain distribution and plastic deformation; and further details were found using sub models. The rubber toughening mechanism was studied by modelling different face-centred micromodels. The stress distributions ahead of the crack tip in global DLJ models were extracted and used as the loading condition for the micromodels, so that a relationship between macromodel and micromodel has been established. It is found that Von Mises and hydrostatic stress play very important roles in the toughening mechanisms and also predicted that rubber particles with multi-layer structure have more potential to toughen epoxy resin than simple rubber particles.

Self-assembly is an effective biomimetic technique for surface functionalization and nanostructural synthesis. 3-Aminopropyltriethoxysilane (APTES) is a popular molecule that can assemble over substrates to modify surface properties. Titanium is a structural material with high weight-specific mechanical properties, corrosion-resistance and bioinertness. Here, a systematic investigation was carried out to optimize the self-assembly of an APTES-modified film on an oxidized titanium surface in order to improve its biocompatibility as an implant material and molecular selectivity, e.g. for CO2 capture. A clean TiOx layer was formed on titanium after the treatment in a Piranha solution of H2SO4 : H2O2 = 3:1. The IR spectra confirmed that the formation of the APTES-modified film (called APS film) on the surface by the presence of the Si-O-Ti and Si-O-Si covalent bonds. The ordering of the self-assembled film did not show strong temperature dependence from 30 to 70ºC, although a thicker film was noted at a higher temperature. Anhydrous toluene as the solvent is essential to the formation of a well-ordered and thin film, compared with hydrous toluene. The well-assembled film was formed on the oxidised titanium surface in the anhydrous toluene solution of ~0.2 v% APTES at 30°C for 16 hours. A higher APTES concentration leads to a disorder film with protonated –NH3 + groups, whereas a lower concentration causes end groups of the adsorbed APTES to loop with the -OH groups on the surface. The APS film with the free –NH2 functional groups is more stable in aqueous solution with pH 10, although it is still hydrolyzed according to the intensity of the –Si-O-Si- bond in the IR spectra. The well-ordered APS film with the –NH2 groups cannot induce heterogeneous nucleation in a simulated body fluid (SBF), because the –NH2 groups are neutral in the solution and the –CH2- hydrophobic groups are exposed in the disordered structure of the APS film. In the application of biofuel cells, the laccase from Trametes versicolor as an enzyme was immobilized on titanium and graphite with the APS film by the covalent bond, respectively. Compared with the native laccase, optimum pH of the immobilized laccase decreased to 3 because of the increase of turnover number (Kcat). Further comparison of Michaelis-Menten constant (Km) of the immobilized laccase with the native one clearly shows that the increase of Km value is mainly due to the change of configuration of the active site, further leading to the lower affinity of immobilized laccase towards the substrate. The laccase on graphite shows higher optimum temperature and twice lower the Km value, compared with the laccase on titanium, which results from the surface morphology of graphite after oxidation. For electrochemical behaviour, graphite with the laccase as electrode does not show direct electron transfer (DET), due to the long electron tunnel between the T1 centre and electrode surface. However, the electrode with laccase shows good mediator electron transfer (MET) in the presence of mediator.

Adjoint based optimisation has until now demonstrated a great promise for optimisation in aerodynamics due to its independence of the number of design variables. This is essential in large industrial applications, where hundreds of parameters might be needed so as to describe the geometry. Although the computational cost of the methodology is smaller than that of stochastic optimisation methods, the implementation and related program maintenance time and effort could be particularly high. The aim of the present is to contribute to the effort of redusing the cost above by examining whether programs using the adjoint methodology for optimisation can be automatically generated and maintained via Automatic Differentiation, while presenting comparable performance to hand derived adjoints. This could lead to accurate adjoint based optimisation codes, which would inherit any change or addition to the relative original Computational Fluid Dynamics code. Such a methodology is presented and all the different steps involved are detailed. It is found that although a considerable initial effort is required for preparation of the source code for differentiation, hand assembly of the sensitivity algorithms and scripting for the automation of the entire process, the target of this research program is achieved and fully automatically generated adjoint codes with comparable performance can be acquired. After applying the methodology to a number of aerodynamic shape optimisation examples, the logic is also extended to higher derivatives, which could also be included in the optimisation process for robust design.

There have been several studies on the tearing of rubber materials since the seminal paper on rupture of rubber was published by Rivlin and Thomas (1953).The behaviour is typically characterised using a fracture mechanics approach whereby the rubber has a geometrically independent relationship between crack growth rate during tearing versus strain energy release rate. This approach works well under conditions of steady tearing as the crack growth rate is easy to measure. However, this approach is much harder to interpret under the condition where the rubber exhibits discontinuous crack growth behaviour such as knotty tearing or stick slip tearing. Unfortunately, these are common tearing conditions observed in practice for filled rubbers as well as for some unfilled rubbers, especially those such as natural rubber that are capable of strain-induced crystallisation. Under these conditions it is not clear what the actual crack growth rate is as the value typically given results from the average of a very rapid tearing rate and a zero velocity tearing rate. The aim of this work is to develop a new approach to characterise the unsteady tearing behaviour of rubber in terms of the relationship between the rate of increase in the strain energy at the crack tip just immediately prior to the onset of the tearing which is quantified directly as the time derivative of the strain energy release rate · T , and the critical strain energy release rate T* required to propagate the crack. The approach adopted in this study is then evaluated using a range of different crystallising and non-crystallising rubbers as well as crystallising rubbers that have been modified to alter their crystallisation over a range of different test temperatures. Additionally, a new elastic-viscous transition diagram in association with the rate of change in the strain energy release rate at the tip of the crack is presented.

The development of remotely addressable tools to encapsulate, store and deliver active materials to living cells is a particularly challenging topic of material science. As drug delivery agents, microcontainers not only require high mechanical stability or to be delivered at target cells, but they should also possess efficient remotely addressable release mechanisms. Light responsive polyelectrolyte capsules are well suited for such purposes. Capsules are constructed using the Layer‐by‐Layer technique where oppositely charged polymers are alternatively deposited on a sacrificial template. The interest for such microcapsules lays in their versatile composition and stimuli‐responsive properties, which can be altered to suit specific needs. The primary aim of this work was to develop polymeric capsules with efficient optically addressable release mechanisms. Previous work on this topic revealed severe flaws in biological environments, especially with regards to the high energy requirements necessary for laser‐induced release and in the very limited knowledge of the fate of microcapsules in living cells. These issues were addressed by developing alternative types of light‐responsive capsules and gaining better understanding of existing ones. Three types of materials were used to sensitize microcapsules to the near‐UV, visible and near‐IR spectral regions: (1) azobenzene‐substituted polymers, (2) gold nanoparticles and (3) photocatalytic porphyrinoid dyes. Various methods were used for the characterization of microcapsules, including laser scanning confocal microscopy, colloidal probe and standard atomic force microscopy, electron microscopy, fluorescence spectrophotometry, UV‐visible spectroscopy and differential scanning calorimetry. Shells were probed for their mechanical stability as well as encapsulation and release behavior based on parameters such as: assembly strategies, shell deformability, permeability, thermal response and response to laser irradiation. This thesis begins with a brief introduction followed by an extensive literature review summarizing the various topics relevant to the work. The materials and methods used in the investigations are catalogued in Chapter 3 . Chapter 4 presents the destructive effects of pulsed UV lasing on polymeric microcapsules and introduces azobenzene‐functionalized capsules with the ability to encapsulate macromolecules by exposure to continuous wave UV light. Chapter 5 looks at the mechanical properties of capsules functionalized with gold nanoparticles as well as their remote release capabilities under near‐IR irradiation. While most of these studies were conducted ex vivo, Chapter 5 concludes with a summary of studies performed in vitro, which demonstrates that it is not only possible to release substances in living cells by light but that the latter also survive in the process. Finally, in Chapter 6, the assembly and light induced destabilization of microcapsules containing porphyrinoid dyes is presented.