Thesis: S.M., Massachusetts Institute of Technology, Department of Materials Science and Engineering, 2014. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 55-61). / Semiconductor nanostructures exhibit distinct properties by virtue of nano-scale dimensionality, resulting in recent interest in semiconducting nanowires for electronic, photonic, and energy applications. Along with nanowires, quantum dots are solution-processable nanocrystals with tunable band gap energies as a function of their size. Based on all of these promising properties that nanostructures exhibit, nanowires and quantum dots are excellent candidates for next-generation optoelectronic devices, including solar cells and light-emitting diodes. However, the realization of nanostructured materials for solar cell device applications is limited by the fundamental trade-off between light absorption and photocarrier collection. Vertically aligned ZnO nanowire arrays can decouple absorption and collection by acting as highly-conductive channels for extracting photogenerated electrons from deep within the film. This thesis illustrates a scheme for the development of ordered bulk heterojunction photovoltaic devices incorporating solution-based n-type doped ZnO nanowires and PbS quantum dots. In order to improve the electrical properties of ZnO nanowires, Al doping of hydrothermally synthesized ZnO nanowires is studied along with the optimization of doping concentration. The morphology of ZnO nanowire arrays is also studied as a function of the doping concentration in the growth solution. Finally, photovoltaic devices are fabricated and the effect of Al-doping of ZnO nanowires is investigated by device characterization techniques. / by Hyoungwon Park. / S.M.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Materials Science and Engineering, 2003. / Includes bibliographical references (p. 245-255). / A study was made that is logically divided into two parts, both involving radiation damage effects. The first is a study of the effects of neutron and gamma radiation on the dimensions of two borosilicate glasses, Pyrex® and Hova SD-2®. These two glasses are commonly used as substrates for silicon microelectromechanical (MEMS) devices, and radiation-induced compaction in a substra.te can have deleterious effects on device performance. Results are presented for density changes induced in both glasses by neutron irradiation. Pyrex was shown to compact at a rate of (in [delta]p[rho]/p[rho] per n/cm2̂) 8.14 x 10-̂20 (thermal) and 1.79 x 10-̂20 (fast). The corresponding results for Hoya SD-2 were 2.21 x 10-̂21 and 1.71 x 10-̂21, respectively. On a displacement per atom (dpa) basis, the compaction of the Pyrex was an order of magnitude greater than that of the Hoya SD-2. Our results are the first reported measurement of irridiation-induced densification in Hoya SD-2. The compaction of Pyrex agreed with a previous study. Our results for gamma irradiations were unexpected. Silicon MEMS strain gauges mounted on glass wafers were gamma-irradiated to hundreds of Mrad. Based on expectations from the literature, the Pyrex was supposed to compact to a level easily measurable by the MEMS strain gauges. Almost no substrate compaction registered in the strain gauges, however. It is hypothesized that the anodic bonding process (by which a silicon wafer was bonded to the glass before etching to create the MEMS strain gauges) was responsible for either 1) changing the bulk radiation response of the glass or 2) creating a layer near the bond interface which somehow prevented the MEMS strain gauges from registering the compaction that was occurring in the glass substrate. While not yet understood, this null result for apparent substrate compaction is of great importance to the problem of mechanically rad-hard MEMS, since it indicates that the response of an anodically bonded Si-glass system to radiation is not simply the sum of the effects on the unbonded materials. To investigate this further, glass samples were prepared in various stages of the anodic bonding process (which involves heating in the presence of an electric field), then irradiated with neutrons. No difference in bulk compaction was noted among the / (cont.) treated samples or the untreated glass, but this result may have been influenced by the high temperature at which the glass was irradiated; however, temperature alone could not have annealed away all the effects of treatment. We conclude that the unexpected results of the MEMS strain gauge experiment were caused by surface layer phenomena at the bonding interface, though we do not currently understand the exact mechanism for this. The second major topic of this study is the effect of neutron irradiation on the Young's modulus of silicon, the constancy of which is key to the operation of many MEMS devices. The elastic constants of defected and amorphous silicon simulation cells were calculated using EDIP. Simulation cells included some containing randomly generated defect distributions, as well as several that were completely amorphous and one containing a small amorphous region. An extensive and careful characterization of point defects was made ... / by Clark L. Allred. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Materials Science and Engineering, 1994. / Includes bibliographical references. / by Ding-Kuo Ding. / Ph.D.

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Materials Science and Engineering, 2017. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 117-128). / The majority of the neural engineering efforts in the past decade have focused on brain interfaces. The searching of tools for recording and modulation of neural activity in the spinal cord limits fundamental understanding of neural dynamics in this organ. Spinal cord poses a challenge to probe design due to its fibrous structure, repeated deformation, low elastic modulus, and sensitivity to implantation procedures. This work addresses the elastic modulus mismatch between spinal cord tissue and synthetic devices by designing flexible multifunctional neural probes capable of conforming to the spinal cord geometry and mechanical properties, while providing functions for optical stimulation and neural recording. In this thesis, fiber drawing techniques are applied to produce flexible and stretchable probes. The utility of the devices for recording and optical stimulation is demonstrated in the spinal cord of transgenic mice expressing the light sensitive protein channelrhodopsin 2 (ChR2). Furthermore, it is shown that the optical stimulation of the spinal cord with the polymer fiber probes induces on-demand limb movements. Finally, the modest dimensions and high flexibility of the devices permitting chronic implantation into the mouse spinal cord with minimal damage to the neural tissue are demonstrated. The findings of this thesis are anticipated to aid the studies of the spinal cord circuits and pave way to new directions in flexible fiber-based optoelectronic devices. / by Chi Lu. / Ph. D.

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Materials Science and Engineering, 2018. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 121-128). / Fibers and fabrics are among the earliest forms of human expression, and yet they have not progressed much from a functional standpoint over the course of history. Recently, a new family of fibers composed of conductors, semiconductors and insulators has emerged. These fibers can achieve device attributes, yet are fabricated using scalable preform-based fiber-processing methods, yielding kilometers of functional fiber devices. Co-draw of different materials is possible for numerous material combinations and sizes, where one of the limiting factors to a continuous feature size in fibers is the Rayleigh-Plateau capillary instability. In my thesis I have shown that it is possible to utilize this adverse fluid instability phenomenon to fabricate uniformly sized and uniformly structured spherical particles internal to the fibers. Judicial choice of the materials and control over the kinetics of this process allowed to integrate the spherical particles into an active fiber device. We have introduced additional ability to control the structure of the fiber by making this process selective, forming high density array of self-assembled spherical photodetectors, connected to continuous electrodes. This fiber structure shows enhanced photoconductivity and sensitivity to wavelength variation, due to spherical geometry of the photoresistive domains. Additionally, an alternative strategy for integrating active devices into fibers was demonstrated. Rather than addressing all the challenges of thermal drawing, we have developed a method to directly integrate commercial functional devices (light emitting diodes, photodetectors etc.) into fibers through thermal drawing. We package these devices internal to the fibers in high density and integrate them with conductive buses, during the thermal draw. This approach enables to combine the benefits of several technologies - high-efficiency devices integrated into kilometer long fibers, which could be weaved into highly functional fabrics. Endowing fibers with active devices will potentially establish a new generation of multifunctional fibers, with highly desired electronic properties. For example, flexible and resilient light emitting fibers could be integrated into textiles to enable covert, optical signal transmission from the soldier uniform to the external world, or high bandwidth photodetectors could be embedded into garments to allow high volume information reception via LiFi (WiFi through light). / by Michael Rein. / Ph. D.

Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Materials Science and Engineering, 1983. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND SCIENCE / Includes bibliographical references. / by Michelle Wendy Gabriel. / M.S.

Thesis (Sc.D.)--Massachusetts Institute of Technology, Dept. of Materials Science and Engineering, 1983. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND SCIENCE / Vita. / Bibliography: leaves 235-242. / by Carol Anne Handwerker. / Sc.D.

Thesis: S.M., Massachusetts Institute of Technology, Department of Materials Science and Engineering, 2014. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 95-99). / Driven by rising environmental and geopolitical concerns, regulations have been put in place over the last decade to compel car makers to lower the CO2 emissions of their cars. Due to these increasingly stringent vehicle efficiency standards, considerable effort has been expended to reduce vehicle fuel consumption. Since the mass of the vehicle dominates all of these efforts, it can be argued that future emission requirements will be impossible to achieve with steel vehicle structures. A transition to lightweight, non-steel materials seems inevitable. However, non-steel materials in most cases require dedicated manufacturing systems due to specific manufacturing requirements. Thus, lightweight vehicle systems will require a distinct divergence between today's manufacturing environment and the potential future manufacturing system. While many studies have assessed greenfield production costs for conventional vehicles and the lightweight alternative, this research recognizes an important reality of the automobile marketplace: any future lightweight vehicle will be implemented out of a steel-based manufacturing environment. Carmakers will have to adapt existing plant infrastructure to the particular requirements of the non-ferrous material. This research develops a conceptual framework and a transition cost model to quantify change penalties of transition processes imposed on vehicle assembly systems. This transition model is applied to a case study provided by Ford Motor Company in order to better understand implications of different manufacturing strategies on the system's capability of switching materials. The research identifies three different manufacturing change penalties which have to be paid when switching the base material in vehicle assembly systems. Taking these penalties into account, case studies suggest when, to what extent, and how materials transitions can be realized most cost-effectively. Partial component-wise transitions are presented as an attractive alternative to full material transitions. Finally, strategies are proposed how to increase the material flexibility of automotive manufacturing systems. / by Christoph Wüstemeyer. / S.M.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Materials Science and Engineering, 2008. / Includes bibliographical references. / Metal nanoparticles hold promise for many scientific and technological applications, such as chemical and biological sensors, vehicles for drug delivery, and subdiffraction limit waveguides. To fabricate such devices, a method to position particles in specific locations relative to each other is necessary. Nanoparticles tend to spontaneously aggregate into ordered two- and three-dimensional assemblies, but achieving one-dimensional structures is less straightforward. Because of their symmetry, nanoparticles lack the ability to bond along specific directions. Thus, the technological potential of nanoparticles would be greatly enhanced by the introduction of a method to break the interaction symmetry of nanoparticles, thus inducing valency and directional interparticle interactions. When a nanoparticle is coated with a mixture of two different ligands, the ligands have been shown to phase-separate into ordered domains encircling or spiraling around the core. Topological constraints inherent in assembling two-dimensional vectors (e.g., ligands) onto a sphere (the core of the nanoparticle) dictate the necessary formation of two diametrically opposed defect points within the ligand shell. The molecules at these points are not optimally stabilized by intermolecular interactions and thus these sites are highly reactive. By functionalizing the polar singularities with a third type of molecule, we generate divalent nanoparticles with "chemical handles" that can be used to direct the assembly of the particles into chains. For example, taking inspiration from the wellknown interfacial polymerization synthesis of nylon, we place carboxylic acid terminated molecules at the polar defect points and join the newly bifunctional nanoparticles into chains by reacting them with 1,6-diaminohexane through an interfacial reaction. / (cont.) Furthermore, we perform a full kinetic and thermodynamic characterization of the molecularly defined polar defect points. We demonstrate that the rate of place-exchange at these points is significantly faster than it is elsewhere in the ligand shell. We also determine the equilibrium constant and standard free energy of the place-exchange reaction at the polar defect sites and demonstrate that the reaction is strongly affected by the molecular environment, i.e. the composition of the ligand shell. / by Gretchen Anne DeVries. / Ph.D.

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Materials Science and Engineering, 2014. / This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. / 127 / Cataloged from student-submitted PDF version of thesis. / Includes bibliographical references (pages 189-199). / With continuously growing energy demands, alternative, emission-free solar energy solutions become ever more attractive. However, to achieve sustainability, efficient conversion and storage of solar energy is imperative. Photoelectrolysis harnesses solar energy to evolve hydrogen and oxygen from water, thereby enabling energy storage via chemical means. Hematite or [alpha]-Fe₂O₃ has emerged as a highly promising photoanode candidate for photoelectrochemical cells. While significant improvements in its performance have recently been achieved, it remains unclear why the maximum photocurrents still remain well below their theoretical predictions. This study investigates the defect chemistry and conduction mechanism of hematite in order to understand and improve this material's shortcomings. A defect model for donor doped hematite was derived and its predictions conformed by the electrical conductivity of ilmenite hematite solid solution bulk samples as a function of temperature and oxygen partial pressure. The enthalpies of the Schottky defect formation and the reduction reaction for hematite were determined as 13.4 eV and 5.4 eV, respectively. In addition, a temperature independent value for the electron mobility of 0.10 cm2/Vs for 1% Ti donor doped hematite was derived. Furthermore, the electrical conductivity of nanometer scale, epitaxially grown thin films of the ilmenite hematite solid solution system was characterized by electrical impedance spectroscopy. This work reports a detailed correlation between the electrical conductivity of the undoped hematite, the 1 atom% Ti doped hematite and the thin films with higher ilmenite content and the conditions under which they were annealed (20° C=/< T =/< 800° c and 10-4 atm =/< po2 =/< atm). Hematite's room temperature conductivity can be increased from ~10-11 S/cm for undoped hematite films by as much as nine orders of magnitude by doping with the Ti donor. Furthermore, by controlling the non-stoichiometry of Ti-doped hematite, one can tune its conductivity by up to five orders of magnitude. Depending on processing conditions, donor dopants in hematite may be compensated largely by electrons or by ionic defects (Fe vacancies). The electron mobility of the film was determined to be temperature independent at 0.01 cm2/Vs for the < 0001 > epitaxial film containing a Ti donor density of 4.0 x 1020 cm-3. Finally, the photoelectrochemical performance of these materials was tested by cyclic voltammetry and measurements of their quantum efficiencies. The 1% Ti doped hematite thin film exhibited the highest photocurrent density of these dense, thin films at 0.9mA/cm2 with an applied bias of 1.5V vs. RHE. The IPCE of this sample reached 15% at wavelengths between 300nm and 350nm after an annealing treatment at 580° for 36 h. The solid solution containing 33% ilmenite preformed nearly as well as the doped hematite. The performance decreased with higher ilmenite concentrations in the solid solution. For all samples containing any ilmenite, the onset potential shifted to lower values by ~200mV after the annealing treatment. The increase in charge carrier density upon reduction of Ti doped hematite was conformed by a Mott-Schottky analysis of the hematite/electrolyte interface. In contrast, only minor changes in the carrier density were observed when reducing an undoped hematite photoanode. Changes in slope of the Mott-Schottky plots revealed the presence of deep trap states in the hematite films. In-situ UV-vis spectroscopy displayed a pronounced optical signature corresponding to the existence of such deep levels. These results highlight the importance of carefully controlling photoanode processing conditions, even when operating within the material's extrinsic dopant regime, and more generally, provide a model for the electronic properties of semiconducting metal oxide photoanodes. / by Johanna Engel. / Ph. D.