A novel electrochemical cell design based on yarn or fiber-like carbon nanotube electrodes was developed as an engineered solution for flexible electrochemical devices, namely electrical energy storage devices. Proof-of-concept cells consisting of porous CNT networks as the sole structural support, electronic conductor and active charge storage material were fabricated and tested. The coaxial yarn cell provided a robust structure able to undergo flexural deformation with minimal impact on the energy storage performance. Greater than 95% of the energy density and 99% of the power density was retained when wound around an 11 cm diameter cylinder. The electrochemical properties were characterized at stages throughout the fabrication process to provide insights and potential directions for further development of these novel cell designs. To demonstrate the ability to improve the performance and extend the applicability of the CNT network based cell design to various cell chemistries and material combinations, the versatile redox activity material vanadium oxide (VOx) was deposited onto CNT yarns. A supercritical fluid deposition and in-situ oxidation process was utilized to create thin conformal coatings of vanadium oxide on carbon nanotube (CNT) surfaces throughout the porous structure of CNT yarns. Half-cell electrochemical characterizations were conducted on carbon nanotube-vanadium oxide (CNT-VOx) yarn electrodes in an 8 M LiCl aqueous electrolyte. The high surface area, interconnected pore structure and high electrical conductivity of the CNT yarn enabled extraordinary rate capabilities from the high capacity Li/VOx system. Cyclic voltammetry tests with scan rates of several volts per second, requiring current densities of hundreds of amperes per gram of electrode mass produced voltammograms with distinguishable redox peaks from Li-ion intercalation/deintercalation. Capacitances of over 150 F g-1 were achieved at a scan rate of 5 V s-1 over a 1.2 V potential window resulting in an energy density of > 32 Wh kg-1 (> 30 Wh L-1) for the yarn electrode. The charge storage also showed good reversibility when cycled over this large potential window, maintaining 90% of the capacitance after 100 cycles at a scan rate of 2 V s-1. Investigation into the structure-property relationship of the CNT yarn and the effects on electrochemical energy storage performance was conducted through half-cell testing of single-filament CNT yarn electrodes. Electrochemical impedance spectroscopy (EIS) enabled the development of an equivalent circuit model based on representation of the CNT yarn through a parallel configuration of a constant phase element (CPE) and Warburg impedance element. Various physical properties of the electrode and electrode-electrolyte system can be obtained from model fitting of the experimental data, providing the ability to verify model accuracy through the comparison of known physical properties (e.g. electrode resistivity), as well as the ability to gain insight into complex electrochemical phenomena from previously unmeasured properties. Analysis of the modeling results for the DC potential dependent EIS tests on electrodes of various lengths suggest a relationship between the effective ion diffusion lengths, the extent of mobile charge carries in the CNT electronic structure and the fractal dimension of the hierarchical self-similar CNT yarn electrodes. The unique concentric cylinder architecture of the flexible coaxial electrode cells displayed an unintended sensitivity of the cell open-circuit voltage (OCV) to the atmosphere surrounding the cell; specifically to changes in the relative humidity. This potentiostatic responsiveness from the coaxial cell was characterized experimentally and a hypothesis was developed relating the OCV response to humidity to the chemical potential energy that exists upon the establishments of concentration (activity) gradients. The Nersnt relation was used to provide quantitative support of the hypothesis through the development of a time-dependent model of the OCV response to the relative humidity. Finite element methods were used to approximate the 1D solution to Fick's diffusion laws to model the H₂O transport through the polymer electrolyte membrane and calculate the time-varying water concentration profile. The model results provide strong support of the hypothesis that the open-circuit voltage response is based on the time-dependent concentration gradient of water between the inner and outer electrodes of the coaxial cell. These results have implications for applications such as humidity sensors, wearable energy harvesting, and self-powered detection devices. / A Dissertation submitted to the Materials Science and Engineering Program in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Fall Semester, 2014. / November 7, 2014. / Includes bibliographical references. / Zhiyong Liang, Professor Directing Dissertation; Tao Liu, Committee Member; Jim Zheng, Committee Member; Steven Lenhert, Committee Member.

A novel ceramic-polymer hybrid composite is fabricated by co-curing lay-up process to combine a carbon nanotube (CNT) reinforced ceramic composite film with a carbon fiber reinforced polymer (CFRP) composite substrate. A transition layer made of non-woven carbon fiber tissues and both ceramic and polymeric matrices are introduced to improve the bonding strength of CNT ceramic composite and CFRP. Volume fraction of CNTs in the CNT ceramic composite is variable from 30% to 60% depending on the CNT preform employed. 6K T300 carbon fiber 2×2 plain weave fabric reinforced bismaleimide (BMI) composites are used as the CFRP substrate. The hybrid composite has good structural integrity with a pull-off bonding strength up to 8.3 MPa. Microstructures are characterized to investigate the bonding mechanism. Ceramic and polymeric matrices are evenly distributed and interlocked each other by the carbon fibers in the transition layer. Carbon fibers in the transition layer bond to the CNT ceramic composite and CFRP tightly with the help of the ceramic or polymeric matrices. Flexural fatigue, heating-cooling thermal fatigue and wet-conditioning test are fulfilled to investigate the structural stability. The hybrid composite developed maintains good properties stability after these tests. Thermal properties of the hybrid composite are studied by both theoretical simulation and experimental method. Micro-hardness test is taken to characterize the surface hardness of the hybrid composite, CFRP and bulk ceramics. Microstructure analysis of the indentation dent confirms that the hybrid composite shares the same mechanical response to micro-hardness test as the bulk ceramics. Sandblasting test in accordance with ASTM C418-12 is applied on both CFRP and the hybrid composite. The CNT ceramic composite layer of the hybrid composite is effective protecting the CFRP substrate from erosions. Microstructure of the sandblasted surfaces is characterized to analyze the failure mode. The erosion resistance mechanism of the hybrid composite is discussed. / A Dissertation submitted to the Materials Science and Engineering Program in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Spring Semester 2018. / April 9, 2018. / bonding, ceramic matrix composite, CFRP, Hybrid composite / Includes bibliographical references. / Cheryl Xu, Professor Co-Directing Dissertation; Eric Hellstrom, Professor Co-Directing Dissertation; Peng Xiong, University Representative; Richard Liang, Committee Member; Zhibin Yu, Committee Member; Chen Huang, Committee Member.

Diblock copolymer self-assembly of materials is emerging as a key element in the fabrication of functional nanostructured materials. By solvent casting or solvent annealing block copolymer thin films, we have demonstrated methods to produce diblock copolymer films with highly oriented, close-packed arrays of nanoscopic cylindrical domains with a high degree of long-range lateral order with few defects. The solvent imparts a high degree of mobility in the microphase-separated copolymer that enables a rapid removal of defects and a high degree of lateral order. Though the use of a selective cosolvent during solvent casting, it was found that the microdomain size and spacing could be increased, leading to a size-tunable system. Additionally, the presence of water also led to the ability to control the microdomain orientation during solvent annealing. Ionic complexation within cylinder-forming PS- b-EO block copolymer thin films was also investigated, where added salts bind PEO block as the minor component. Small amounts of added salts, on the order a few ions per chain, show large effects on the ordering of the copolymer films during solvent annealing. By using gold or cobalt salts, well-organized patterns of nanoparticles can be generated in the copolymer microdomains. Topographically and chemically patterned surfaces were used as a route to sectorizing and controlling the lattice orientation of copolymer films. Topographically patterned surfaces allow well-defined boundaries to confine the copolymer microdomains on a surface and effectively direct the ordering and grain orientation of the copolymer microdomains. Chemically patterned surfaces provide a route to direct the block copolymer ordering on completely flat surface, which may have advantages in applications where adding additional topography may be undesirable. To generate nanoporous templates from PS-b-PEO bases materials several routs were followed. The first route was through the addition and selective solvent removal of homopolymer PEO or PMMA. Second, we have incorporated a center block that is photodegradable by ultra violet radiation into PS- b-PMMA-b-PEO copolymers. Third, a tritylether junction was placed between the two blocks, which is cleavable by exposure to trifluoroacetic acid vapor. Though the use of solvents in block copolymer thin films, were are able to markedly enhance the long range lateral ordering block copolymer films. Also, routes to sectorize surfaces to confine and direct the copolymer microdomains are shown. Also, three methods to generate nanoporous films from PS-b-PEO based copolymers are demonstrated. All of these results are important in the realization of addressable media from block copolymer nanolithography.

Includes bibliographical references / Two magnesium/silicon aluminium alloys each reinforced with 20 vol. % SiC particulates have been worn against three different automobile friction linings (brake pads). Two of the friction linings are commonly used against cast iron brake rotors while the third has been formulated for use against aluminium MMC brake rotors. Wear processes at the interfaces of the specific rotor - pad combinations have been characterised through the analysis of friction traces and the use of optical and electron microscopy. Models on the interdependence of friction and wear, and models of wear mechanisms for aluminium MMCs and cast iron sliding against friction materials have been proposed and discussed. For an aluminium MMC sliding against an organic pad formulated for use against cast iron, wear rates are low and friction is constant due to the formation of a solid lubricant layer at the wear interface. When this MMC is worn against a semi-metallic pad formulated for use against cast iron, wear rates are extremely high due to two and three body abrasion which lead to subsurface delamination and early melt wear in the MMC. For an aluminium MMC developed for its use in automobile brake rotors sliding against a semi-metallic pad specifically formulated for its use against MMC brake rotors, wear rates at low loads are low although friction traces are irregular and fracture of the SiC particulates occurs at the lowest load and sliding velocity. This fracture of SiC is caused by the abrasive action of hard alumina particles within the pad. At high loads and sliding velocities cohesiveness of materials within the pad is poor and the wear rate of the MMC is extremely high. At the highest load/sliding velocity combination, the wear resistance of the MMC is inferior to that of its unreinforced matrix. If the structure and composition of friction linings are arranged correctly, the wear resistance and frictional performance of aluminium MMC brake rotors are superior to those of cast iron brake rotors. In addition, the lower density of aluminium MMCs provides for an economic advantage over cast iron with respect to efficient use of fuel, and fabrication expenses.

Friction plays an intricate role in our everyday lives, it is therefore critical to understand the underlying features of friction to better help control and manipulate the response anywhere two surfaces in contact move past each other by a sliding motion. Here we present results targeting a thorough understanding of soft material friction and how it can be manipulated with patterns. We found that the naturally occurring length scale or periodicity (λ) of frictionally induced patterns, Schallamach waves, could be described using two materials properties (critical energy release rate Gc and complex modulus (E*), i.e. λ∞ Gc /E*). Following this, we evaluated the effect of a single defect at a sliding interface. Sliding over a defect can be used to model the sliding from one feature to another in a patterned surface. Defects decreased the sliding frictional force by as much as 80% sliding and this decrease was attributed to changes in tangential stiffness of the sliding interface. The frictional response of surface wrinkles, where multiple edges or defects are acting in concert, was also evaluated. Wrinkles were shown to decrease friction (F) and changes in contact area (A) could not describe this decrease. A tangential stiffness correction factor (fx) and changes in the critical energy release rate were used to describe this deviation (F ∞ Gc *A*fx/ℓ, where ℓ is a materials defined length scale of dissipation). This scaling can be used to describe the friction of any topographically patterned surface including the Gecko's foot, where the feature size is smaller than ℓ and thus replaces ℓ, increasing the friction compared to a flat surface. Also, mechanically-induced surface defects were used to align osmotically driven surface wrinkles by creating stress discontinuities that convert the global biaxial stress state to local uniaxial stresses. Defect spacing was used to control the alignment process at the surface of the wrinkled rigid film/soft elastomer interface. These aligned wrinkled surfaces can be used to tune the adhesion and friction of an interface. The work presented here gives insight into tuning the friction of a soft polymeric surface as well as understanding the friction of complex hierarchical structures.^

Abstract not available

The post-yielding behavior of some common thermoplastics was examined in uniaxial tension to determine if these materials were ideally plastic from a thermodynamic viewpoint. Various polyethylenes, poly(methyl methacrylate) and polycarbonate, polyarylate and polysulfone based on bisphenol A were studied. Thermodynamic measurements were made during deformation using a novel isothermal deformation calorimeter capable of measuring the work and heat of deformation. Thermodynamically ideal plasticity was not observed for any of the polymers examined. The polyethylenes stored approximately 30% of the input work as a latent internal energy change while this value was 40-50% for the amorphous glasses. Differential scanning calorimetry results for the deformed polyethylenes indicated that the heats of transition were less for the drawn samples than for the isotropic samples. This result was primarily due to the stored deformation energy and was not necessarily indicative of a change in crystallinity. The energy stored during drawing was explained using some commonly accepted models for the deformation of polyethylene. Additional experiments were performed to determine the mechanism of deformation energy storage and to ascertain the implications of this stored energy in engineering applications. Relaxations at temperatures much less than the glass transition temperature (T$\sb{\rm g}$) were observed for the drawn amorphous glasses using dynamic mechanical and differential scanning calorimetry measurements. Substantial thermal shrinkage was found in unconstrained drawn glassy samples exposed to thermal cycles never exceeding T$\sb{\rm g}$. Considerable stress buildup was also observed for uniaxially constrained glassy samples cycled at temperatures much less than T$\sb{\rm g}$. The values of these stresses were typically greater than 50% of the yield stress. These thermally induced events were attributed to a partial release of the energy stored during deformation. These experimental observations were explained in terms of a proposed deformation model which involves chain deformation with breakage and re-formation of intermolecular secondary bonds.

The thermodynamics of uniaxial solid deformation was studied experimentally for a number of polymeric solids, including two polyurethane-urea elastomers, natural rubber, a thermoplastic elastomer and low density polyethylene. A deformation calorimeter was developed to measure the heat and work of uniaxial solid deformation and measurements were made on the above materials. A differential scanning calorimetry method was developed to characterize the melting behavior of stretched elastomers which were found to undergo stress-induced crystallization during stretching as deduced from the large but recoverable internal energy changes measured by deformation calorimetry during uniaxial extension and contraction. Wide angle x-ray diffraction and thermostatic measurements were also performed on the elastomers held in the extended state in order to characterize the amorphous-crystalline phase transition which occurs during deformation. The motivation for this work was to evaluate the performance of the two polyurethane-urea elastomers which were found to function effectively as working substances in rubber heat engines. These elastomers could generate 1 Joule of work per gram of elastomer at about 3% of Carnot efficiency in experimental Sterling cycles.

The major theme of this dissertation is structure/property relationships in fiber-reinforced thermoplastic-matrix composites. Effort has been focused on the interface: interfacial crystallization and fiber/matrix adhesion. Included are investigations on interfacial nucleation and morphology, measurement of fiber/matrix adhesion, effects of interfacial adsorption and crystallization on fiber/matrix adhesion, and composites reinforced with thermotropic liquid crystal copolyester fibers. Crystallization of a copolyester and poly(butylene terephthalate) with glass, carbon, or aramid fibers has been studied with regard to interfacial mophology. Techniques employed included hot-stage optical microscopy and differential scanning calorimetry. Nucleation by the fibers was found to be a general phenomenon. Morphology could be varied by changing the cooling rate. In order to better monitor fiber/matrix adhesion, a buckled plate test has been developed. The test measures transverse toughness as the parameter characterizing interfacial adhesion in unidirectional, continuous-fiber composites. The test is simple to perform yet has advantages over other interfacial evaluation techniques. The buckled plate test was found to be a sensitive measure of fiber/matrix adhesion. The buckled plate test has been used along with the transverse tensile test to examine how interfacial adsorption and crystallization affect fiber/matrix adhesion in polycarbonate/carbon fiber composites. Adsorption was found to be of primary importance in developing adhesion, while crystallization is a secondary effect. The toughness data have been fit successfully for annealing time and temperature dependence. The dependence of adsorption and transverse toughness on matrix molecular weight was found to be large, with higher molecular weights adsorbing more effectively. Studies of the fiber/matrix interface have been extended to composites reinforced with thermotropic liquid crystal copolyester fibers. Composites made with these fibers had poor transverse properties, regardless of matrix. Surface treatment such as ozonation increased transverse properties, but values were still low. Scanning electron micrographs of fracture surfaces indicated that fiber splitting occurs, especially for surface treated fibers. Poor fiber transverse properties rather than fiber/matrix adhesion thus appear to limit composite transverse properties.

Defects in semiconductors can play a vital role and even dominate the performance of optoelectronic devices. Thus, understanding the relationship between structural defects and optoelectronic properties is central to the design of new high-performance materials. In this dissertation, we apply state-of-the-art first-principles approaches based on density functional theory (DFT) and many-body perturbation theory (MBPT) to quantitatively describe trap state energies and optical excitation spectra of defective bulk gallium nitride (GaN) and monolayer germanium selenide (GeSe). GaN is a technologically important wide bandgap semiconductor used as a power electronics and blue light emitting material, and naturally contains performance-degrading defects. For GaN containing a charged nitrogen vacancy, we systematically study the trap-state energies and excitonic properties. We benchmark the accuracy of hybrid DFT by comparison to MBPT studies of defective bulk GaN and determine that the HSE functional (Heyd–Scuseria–Ernzerhof) predicts trap-state energies in excellent agreement with MBPT, and that a recently developed solid-state screened range-separated hybrid (SRSH) functional can quantitatively reproduce MBPT-predicted defect energetics, including optical excitations. Additionally, we utilize MBPT to quantify the localization of the Wannier-Mott exciton in the presence of a point defect, introducing an analysis technique of the exciton envelope and center-of-mass functions to extract the Wannier exciton Bohr radius and quantify the perturbation of the exciton wavefunction due to the defect. We then utilize (TD)SRSH to study the excited-state properties of three other important defects in GaN and predict that the carbon impurity may result in the well-known yellow luminescence in bulk GaN. Finally, we apply MBPT with the same analysis techniques developed for GaN to study the optoelectronic properties of defects in monolayer semiconducting GeSe, a material that has promising applications in next-generation optoelectronic devices; we determine that a selenium vacancy strongly modifies the optoelectronic properties of the material. Overall, this dissertation provides a recipe for performing quantitatively accurate MBPT and TDDFT calculations on defective semiconductors, with a systematic study of calculation convergence and defect-defect interactions. Additionally, by an analysis technique of the BSE-computed exciton wavefunction, we introduce a framework for describing defect-induced exciton localization that can be broadly applied to many classes of materials.