RSS 1.02022 - Mines Theses & Dissertations
Browse by
Recent Submissions
-
Near-real-time/real-time methane prediction and visualization in underground longwall coal miningPreventing methane explosions remains a challenging task in underground longwall coal mines. Mine-wide real-time methane monitoring is a must to tackle this challenge. Real-time decision-making is critical to stop possible explosive methane accumulation and prevent accidents. To accomplish this, detecting and monitoring methane emissions and concentrations across the entire mine must be predicted. Atmospheric monitoring systems are a critical part of monitoring mine ventilation systems in real-time, and the number and location of sensors are essential to detecting any possible hazard. However, sensors can not be installed in the most critical locations, such as the shearer cutting drum and along the longwall face, due to the nature of mining equipment’s movement. Thus sensors provide information for a limited area, and their readings may have delays caused by the sensor response time, gas diffusion rate, and temperature. Computational fluid dynamics modeling can provide relatively accurate predictions regarding the location of possible explosive gas concentrations; however, it requires significant computational resources and time, which is not conducive to real-time decision-making. Lastly, decision-makers (mining personnel) need to observe and assess predictions to take the required precautions. Although automated shutdown mechanisms can be used in the event of a sudden increase in methane emissions, these are not entirely reliable due to previous accidents. Such mechanisms might also interfere with productivity as a result of unnecessary stops, lost time, and decreased production. Therefore, the predictions need to be visualized in a way that offers the fastest and most accurate response time for personnel. This dissertation explores, evaluates, and benchmarks suitable artificial intelligence (AI) algorithms for predicting 3D near-real-time explosion hazards. The prediction performance of 10 algorithms is compared using seven datasets and then assessed through classification accuracy. With the best performer chosen, an application-specific methodology is proposed with modified long short-term memory to detect the formation of explosive methane-air mixtures in the longwall face and identify possible explosive gas accumulations before they become hazards. Lastly, the visualization of the predictions is compared in three platforms, namely computer screen, virtual reality (VR), and augmented reality (AR), for accurate methane volume assessment and faster response time. The best-performing algorithm, Long Short-Term Memory (LSTM), is modified, trained, tested, and validated based on CFD model outputs for six locations of the shearer for similar locations and operational conditions of the cutting machine. The results show that the algorithm can predict explosive gas zones in 3D with overall accuracies ranging from 87.9% to 92.4% for different settings. Depending on the prediction area, output predictions take between 30 seconds to two minutes after measurement data are fed into the algorithm. Lastly, a user study is designed to investigate the effects of two visual variables, transparency/opacity and complex/simple, to compare different platforms. Four simulations are tested with 30 human subjects. Participants are asked to determine the explosive gas volume. The subjects` time of completion and eye-tracking data are also recorded. Along with the recordings, saliency maps of four simulations are created. Recorded data are evaluated with both quantitative and qualitative analysis. The quantitative analysis assesses the accuracy of explosive gas volume detection and completion time. In addition, the saliency maps of the designed simulations with the subject's eye-tracking data is analyzed to investigate the subjects' attention and reaction in AR, VR, and computer screen. The qualitative analysis assesses subjects` responses to questions about their feelings. The results of the quantitative analysis show a statistically significant difference between the computer screen and VR for the transparent and simple scenario. The subjects` accuracy of volume determination is better in VR. Moreover, the eye tracker data show that subjects spent more time in non-salience regions in VR than on the computer screen. In addition, the qualitative analysis reveals that 90% of the subjects preferred to use VR due to the immersive experience, and 67 % felt more confident in VR when investigating the explosive zones and flagging them as dangerous zones. Finally, the users’ preference ranking of platforms for explosive zone estimation is VR, computer screens, and AR.
-
Extraction and simulation of nonlinear microwave devices using the finite-difference time domain methodComputer simulation of electromagnetic behavior is a key step in the design cycle of many electronic devices, including microwave communications components and high-speed digital devices. One widely-used simulation technique is the finite-difference time-domain (FDTD) method, wherein Maxwell’s equations are simulated in the time domain within a three-dimensional grid. A complicating factor for electromagnetic simulation is the incorporation of nonlinear components such as diodes and transistors. These components require specialized models, which add complexity to the simulation software. Versatile models, such as the X-Parameters, are attractive because a single nonlinear model can encapsulate any number of nonlinear devices simply by changing the model parameters. In this work, methods for incorporating the X-Parameters with FDTD simulation are developed. The X-Parameters model captures nonlinear behavior in the form of harmonic interactions. A method for extracting X-Parameter information by post-processing recorded data from an FDTD simulation is described. This method is tested on a nonlinear diode, and results compared to those from a commercial simulation package with an internal X-Parameters implementation. The X-Parameter model was additionally incorporated into FDTD simulation by means of a “black-box” updating formulation, which allows an FDTD grid cell to encapsulate the behavior of a circuit component with known X-Parameters. This X-Parameter black-box was tested for a bipolar-junction transistor, and results compared to established simulation approaches in FDTD. This novel approach is a versatile means for incorporating nonlinear devices into the FDTD simulation method.
-
Geochemistry of the Niobrara Formation and oceanic anoxic event III (OAE III): a regional study in the Western Interior SeawayThe Coniacian – Santonian (C – S) time interval displays positive enrichment of stable carbon isotopes (CIE) globally, and elevated organic carbon burial under oxygen depleted conditions in the Western Interior Seaway (WIS) and proto-equatorial Atlantic regions. The CIE associated with C – S has been described as the Oceanic Anoxic Event III (OAE III) with a duration of approximately 3 my and high organic carbon potential. The OAE III in the WIS correlates to the Niobrara Formation C marl and C chalk intervals and Scaphites Depressus biozone in the Denver Basin and is represented by the Razor 25-2514H core (Redtail Field) in this study. Redox sensitive trace elements (molybdenum (Mo), vanadium (V), and iron (Fe)) indicate a range of redox conditions from dysoxia to persistent euxinia were present during the regional OAE III in the Denver Basin. Water-column stratification and deep-water mass restriction were established and displayed varying degrees at the onset and during the event. The OAE III is continuous across the WIS based on redox sensitive Mo trends indicating positive enrichment, except in the Piceance Basin. The analysis of the high-resolution bulk sedimentary XRF results indicate three distinct anoxic phases which are informally named OAE IIIa, OAE IIIb, and OAEIIIc. OAE IIIa and OAE IIIb displays the strongest anoxic conditions, while OAE IIIc marks the waning period of anoxia in the WIS. Each of the subdivisions are separated by intervals that display low enrichments of redox sensitive trace elements that indicate periodic oxygenation of the water column in between anoxic phases. Based on the acyclic, saturates, and aromatics biomarkers results of the Niobrara Formation, the organic matter (OM) is primarily composed of algae remains and terrestrial woody material. The relative concentration of the algal and woody organic matter depends on the paleogeography where western margins display more woody organic matter content whereas eastern parts show more algal input. The OM in the Niobrara Formation was deposited in epeiric seaway conditions. Ocean salinity levels in the WIS differed from the Gulfian Regions based on the distribution of Oleanane (Ol/H) and Moretane (Mo/H). Gammacerane (Ga/H) index indicates anoxic conditions were stronger in the Powder River Basin than the Cañon City Embayment and the Piceance Basin. The presence of tricyclic terpanes is either related to presence of Tasmanites algae or formation of tricyclic terpanes through cracking of tetracyclic terpanes. Dinosterane index readings from the Cañon City Embayment and Piceance Basin indicate that dinoflagellates were not present in the western parts. In the OAE III, trends in the relative abundances of n-alkanes, pristane and phytane indicate the organic matter composition changed and became more algal compared to the Fort Hays member. Source rock analysis (SRA) results show substantial increase in hydrogen index (HI) and decrease in oxygen index (OI) values at the onset of the OAE III supporting an increase in algal contribution to the OM composition as well as formation of oxygen depleted conditions. During the OAE III, phosphorus (P), barium (Ba), cadmium (Cd), nickel (Ni) and copper (Cu) were recycled and led elevated production rates. Ni and Cu trends with low values in the OAE IIIa and OAE IIIb, indicate lack of free hydrogen-sulfide (H¬2S) in the water column. Elevated amounts of Ni and Cu deposition took place during the OAE IIIc. A sharp positive increase in the concentrations of redox sensitive trace elements that correlate to the B chalk, A marl, and A chalk intervals indicate another episode of oxygen depletion during the deposition of Smoky Hill Member under high marine productivity rates. Silicon (Si) and titanium (Ti) indicate major changes in the paleoclimate that occurred with the onset of OAE III. Dry climate conditions transitioned to more wet/humid climate promoting elevated continental weathering rates and subsequently larger quantities of nutrient transportation to the WIS. Potassium (K) and magnesium (Mg) show a period of steady stream influx followed by quiescence at the onset of the OAE III. The low riverine influx rates are due to the initial deepening of the WIS which is followed by an increase in the stream influx rates during the deposition of the C marl interval. Paleoproductivity, nutrient recycling, paleoclimate, and riverine influx trends across the WIS show drastic changes associated with the paleogeography. In the western margins, trends are cyclic due to the proximity to sediment source areas whereas in deeper parts of the WIS, the trends better show influence of anoxia on trace metal sequestration. Overall, the results from this study indicate that the OAE III in the Denver Basin include three discrete anoxic intervals and was initially formed during the deepening of the WIS above the Fort Hays member, and was prolonged due to the elevated production rates that took place under wet/humid climate conditions and varying degrees of riverine influx during the C marl and C chalk intervals. The OM composition changed from more terrestrial to more marine algal at the onset of the OAE III and persisted throughout the Niobrara Formation. Biomarker results provided important knowledge on the WIS-wide changes in the OM composition. The regional correlation of the OAE III shows the paleogeography of the study sites were the determining factors in terms of anoxia versus oxygen rich conditions. The paleoproductivity, nutrient recycling, riverine influx rates, and paleoclimate varied across the WIS.
-
Resource estimation of roll front uranium deposits by using traditional and machine learning methods for Nichols Ranch uranium deposit in WyomingThe fundamental component of resource estimation is domain modeling. However, in such deposits like roll front uranium is deposited in sandstone and on the contact of oxidized and reduced rock domains which makes it difficult to model. Grade - Thickness (GT) contour method is one of the most applied resource estimation techniques in roll-front uranium deposits. However, explicit modeling and GT contouring by using the GT information extracted by drill holes is exceedingly difficult, time consuming and inconsistent. This thesis studies a domain modeling of roll front uranium mineralization using the GT values within the radial basis function (RBF) aided implicit modeling framework. And compares the spatial associations of the RBF determined domain of mineralization to the domains obtained by GT contours. Then, the block grades within the domain of mineralization are estimated by using Kriging and selected machine learning models to compare their spatial associations and overall, in-situ tons and grade estimates with each other and GT contours. The domain of mineralization modelled by RBF appears to spatially correlate well with the domain of mineralization obtained by GT contours. The performance of the selected machine learning models was quite good with k -NN having values of R2=0.792, RMSE=0.0216, and MAE=0.0048 and the random forest with R2=0.751, RMSE=0.0236, and MAE=0.0077. A visual validation of these models, swath plots, grade tonnage curves suggests that the k -NN and Ordinary Kriging (OK) results are remarkably close to each other perfectly aligning with the drillhole intersections in terms of grades while random forest (RF) estimates show significant deviations of higher grades from the other methods and the supporting drill hole information. For final comparisons, it was observed, in the study area that the difference between GT in-situ resource estimates and OK and k -NN results were approximately 4% and 1% respectively.
-
Integrated multiphase flow modeling in wellbore for downhole pressure predictionsThroughout the life of the well, when designing many aspects of various disciplines that pertain to the flow of fluids within a conduit, knowing the bottomhole flowing pressure (P_wf) and pressure profile in the wellbore are of undeniable significance. One would ask why not deploy downhole pressure gauges to obtain P_wf readings. Unfortunately, it is not economical or practical to do so. The harsh in-situ conditions are not optimal for such gauges, resulting in high failure rates. Thus, the common practice is to apply hydraulic models to predict P_wf given surface measurements. To do so, predicting the fluid flow behavior is vital. The prediction of such behavior is quite simple when dealing with single-phase fluid flow. Unfortunately, such is rarely the case in the petroleum industry. The existence of multiple phases introduces multiple complexities hindering the accuracy of the predictions. Even the most sophisticated models pertain to a considerable amount of error when applied to some conditions. To solve this issue, comprehensive mechanistic models that integrate fluid flow mechanisms to vastly increase the accuracy of the predictions could be used. Although the mechanistic models can be considered the most accurate multiphase flow model, they cause major difficulties when coupled with simulators due to their discontinuous nature. In this work, a new integrated multiphase flow model was developed. The aim of the new model is to produce relatively accurate downhole pressure predictions under all flow conditions while maintaining a simple, differentiable, and continuous form. In this study, the performance of widely used multiphase flow point models has been evaluated with five different field datasets. Then the integrated model was put together by incorporating the state-of-the-art onset of liquid loading model. Unlike the existing models, the new model classifies the flow based on the onset of liquid loading. If liquid loading occurs, then the model uses a sophisticated drift-flux model which proved to be both simple and relatively accurate when dealing with high liquid flow. On the other hand, if liquid loading does not occur, an improved two-fluid model will be used. The proposed two-fluid model was developed by improving on the wetted perimeter and the liquid wall shear stress of existing two-fluid models. The proposed model outperformed five other models in predicting the liquid holdup and pressure gradient of 11 experimental datasets (1478 data points). In addition, the model succeeds capturing the effects of inclination angle, gas density, and liquid and gas superficial velocities on liquid holdup and pressure gradient. With the improved two-fluid model, the new integrated model was completed. Moreover, the new integrated model outperformed the other multiphase flow models in predicting the downhole pressure of 313 field data points. When dealing with wells that have both segregated and intermittent flow, the new model produces outstanding results. In addition, it succeeds in finding the location and amount of liquid loading in a well. Ultimately, the new integrated model paves the way for the design and optimization of artificial lift processes that require knowing the location and amount of liquid loading.
-
Rare earth reduction: a techno-economic overview of the state-of-the-art technology and novel developmentsRare Earths are critical to the future of green technology, especially for wind turbines and electric vehicle motors. However, reducing these rare earths into their metals is challenging metallurgically and economically. An overview will be given on the state-of-the-art processes for reducing rare earths from the rare earth oxides to the rare earth metal. Included are some of the novel processes being developed in the field. The market for these processes is examined at an international level. Also, the capital and operating costs of the state-of-the-art processes, along with some of the novel processes, were examined. This lays the foundation for the direction of future trends in commercial rare earth reduction technology.
-
Finite element analysis of the laser perforation process: thermal and mechanical perspectivesA great challenge to using lasers for the purpose of drilling or perforating lies in understanding the complex processes occurring simultaneously and involving thermal, mechanical, and photonic changes. As a means to better such understanding, a fully coupled thermal-mechanical finite element model was developed to investigate the ensuing thermal and mechanical effects due to laser perforating a sandstone rock sample. The model was validated against experimental data based on the perforating rate and the created perforation profile. Eight sensitivity studies were conducted to understand the separate effects of the laser perforating process parameters. Additionally, a steel-rock model was created to simulate laser perforating in cased-hole completions. In all simulated cases, two distinct stress regions were formed as a result of laser perforating. The first one is the hotter lased region which was under compressive stresses opposing the induced thermal expansion, and the second one is the colder surrounding region, which was under tensile stresses to contain the thermally expanding hotter region. The sensitized process parameters encompass laser beam power, laser beam radius, confining stresses, Young’s modulus, specific heat capacity, thermal conductivity, thermal expansion coefficient, and vaporization temperature. It was found that the perforating rate was directly proportional to the laser beam power and inversely proportional to the laser beam radius. Altering the magnitude of confining stresses, confining stresses ratio, Young’s modulus, or thermal expansion coefficient, did not change the perforation rate or profile, albeit influencing the magnitude of the thermally induced stresses. A rock with a larger specific heat capacity was harder to be heated, thus was under a slower perforating rate. Also, greatly increasing the rock’s thermal conductivity resulted in somewhat a slower perforating rate. Furthermore, increasing the vaporization temperature to more than double the base value also slowed the perforating rate. Overall, changing the rock’s thermal properties or laser design parameters affected the ensuing perforating rate and profile, whereas altering the rock’s mechanical properties only caused changes to the generated thermal stresses. Compared to only 30 seconds to perforate the simulated sandstone rock sample, laser perforating both steel (top) and rock (bottom) layers lasted 50 seconds. The first steel element was removed after 10 seconds of laser heating, whereas removing the first rock element took approximately 18 seconds. Simulation results showed that steel was harder to laser perforate than rock, mainly attributed to steel having a larger melting/vaporization temperature and a significantly greater thermal conductivity. The stress profiles obtained in the stress-rock model were similar to the rock-only model, where compressive stresses are dominating the hotter lased area and tensile stresses are distributed in the colder surrounding area. However, the concentration of these thermally induced stresses is more pronounced in the steel layer than in the rock.
-
Advanced transmission electron microscopy characterization of complex nanomaterials for hydrogen storage and conversionIncreasing energy demands and high carbon emissions require society to move toward renewable energy sources like hydrogen; however, the optimization and cost reduction of nano-scale materials in hydrogen energy systems is critical to implementing a hydrogen-based energy economy. In hydrogen power generation, fuel cells require high amounts of costly catalyst materials, which have limited durability and reduce the cost efficiency and lifespan of the system. In parallel, hydrogen storage technologies would benefit from the development of materials that transport hydrogen at higher densities than current compression technologies. This work aims to fill these gaps through investigations of the effects of nitrogen doping in carbon supports for fuel cells as well as introducing and characterizing a novel hydrogen storage material. Nitrogen-doped carbon supports have shown improvements in the catalyst-support interactions for fuel cell catalysts, but their inherent heterogeneity limits our ability to tune their performance. In the first study, the effect of pyridinic and graphitic nitrogen on metal nucleation is investigated. This work begins with computational analysis to assess the absorption energy and electronic occupation of twelve metals across the periodic table in the presence of pyridinic and graphitic nitrogen defects. The computations inform the down selection of metals to test experimentally, where non-platinum-group metals (Fe, Co, and Ni) are synthesized on pyridinic-rich and graphitic-rich nitrogen-doped carbons. Low-voltage scanning transmission electron microscopy (STEM) is then used to determine the extent of metal nucleation for Fe, Co, and Ni. Experimental and theoretical trends were then compared and demonstrated trends that can be traced back to the electronic structure of the metal and its interaction with the nitrogen defect. Next, this dissertation investigates the effect of the same nitrogen-doped materials on the dispersion and stability of pre-formed Pt nanoparticles. Platinum nanoparticles are deposited using a polyol method onto four chemically-varied, nitrogen-doped carbon supports. Initial dispersion of the platinum particles is evaluated using high-resolution STEM. The resulting electron micrographs are analyzed using a machine-learning-based technique to identify and quantify the size of thousands of platinum particles, increasing the quantity of data from hand-selection methods. The stability of the platinum catalyst is then probed through electrochemical stress testing and imaged at various points in the test using identical location STEM. A combination of machine-learning-based image analysis and manual identical location analysis suggests better dispersion and stability of pre-formed Pt nanoparticles on nitrogen-doped carbons with a slightly higher percentage of graphitic N defects. Lastly, a new, highly energetic material that allows for the rapid release of hydrogen is synthesized and characterized. A new set of protocols for the vapor phase addition of TiCl4, BBr3, and N2H2 to Mg(BH4)2 results in a new hybrid composite material that releases an impressive 7.6 wt % of hydrogen at temperatures as low as 100 °C. Differences in composition and morphology between the series of modified samples are demonstrated using identical location microscopy paired with other physical and chemical characterization techniques. This investigation determines how the modification of Mg(BH4)2 resulted in the presence of N2H5Br and N2H5Cl, which attributes to the mechanism of low-temperature hydrogen dehydrogenation. This work also spotlights the benefits and caveats of utilizing identical location microscopy for metal hydrides in the hydrogen storage field due to their air- and beam-sensitivity. The final chapter summarizes the findings seen in the discussed studies and proposes the suggested directions for future work.
-
Poroelastic properties of carbonates: a case study on Indiana limestonesThis thesis studies the poroelasticity of carbonate rocks and investigates the dependence of various physical properties on confining pressure and pore pressure. These physical properties include porosity, permeability, static strain, ultrasonic P- and S-wave velocities. In a process, I estimate dynamic and static stiffnesses of carbonates as they often differ due to various effects, such as velocity dispersion, heterogeneity of the rock, and amplitudes of applied strain. The studied materials include carbonate rocks with similar mineralogy and porosity, but varying permeability to see the effect of hydraulic properties on the elastic properties. The samples with higher permeability had larger stiffness values and faster velocities, which I explain through a lower specific surface area per grain volume and hence, better cementation of grains. Estimated elastic properties were then used to produce dynamic and static effective stress coefficients, which were found to have a linear relationship. I then characterize the effective stress law for strain for an Indiana limestone rock by measuring the volumetric strain as function of confining and pore pressures. Biot’s coefficient decreased from 0.70 to as low as 0.59 in response to a 35 MPa change in differential pressure. Such a strong pressure sensitivity implies the importance of accurate estimations of ? with respect to pressure rather than using a constant value. Finally, I experimentally demonstrate the interaction of the rock with deionized water as a pore fluid; during exposure to aggressive intrinsic fluid, the rock’s solid matrix becomes softer. This type of rock-fluid interaction affects fluid substitution models, such as Gassmann’s model, which fail to predict the actual stiffness of the saturated rock.
-
Multiscale characterization of defects in 4H-SiC high-power devices and the effect on MOSFET reliabilitySilicon carbide (4H-SiC) has been accepted as an optimal semiconductor that can substitute for silicon for fabricating advanced power devices for high temperature, power, and frequency applications, owing to its outstanding physical properties. Despite the major progress in SiC process technology, SiC-based devices suer from bipolar degradation. As an example, SiC p-i-n diodes have previously suered from an increase in the forward voltage drop under forward conduction stress. It was discovered that the basal plane dislocations (BPDs) in epitaxial layers result in the formation of stacking faults (SFs) that can expand through a mechanism called \recombination enhanced glide mechanism" whenever the p-i-n diode is forward biased. The SFs represent regions of poor lifetime and are regions with poor conductivity modulation. Indeed, SFs not only act as recombination centers but also impede the ow of majority carriers. The occurrence of SFs must therefore be prevented, the expansion behavior needs to be understood and correlated to SiC-based device performance. For this reason, the central focus of this dissertation is to provide a full understanding of specic types of extended defects found in commercial wafers through a multiscale analysis and study the eect on 4H-SiC metal{oxide{semiconductor eld-eect transistors (MOSFETs) reliability. The rst study highlights the power of a multiscale luminescence characterization approach to studying extended defects in epitaxial 4H-SiC semiconducting materials using two complimentary techniques, photoluminescence (PL) and cathodoluminescence (CL). The BPD network generated from strain around a down-fall particle indicated the presence of dierent structures, such as Shockley-type and Frank-type SFs. Ultraviolet-PL imaging allowed for a rapid identication of the inner structure of the defect by revealing the BPD network and the presence of various SFs, and CL was used to provide better spatial and spectral information. This detailed optical analysis provides a pathway for the fundamental understanding of the impact of defects on device performance and provides a better understanding of their formation and development during epitaxial growth. In this work, an inclusion was selected as an example, but this method applies to any heterostructures or areas that show BPDs. The nature, origin and behavior under device operation of so called "trapezoidal defects" were revealed using a complete multiscale characterization study and the results were correlated with degradation of MOSFETs having this defect. The correlation between the luminescence and microscopy results allowed us to precisely identify the nature of these SFs as: Single Shockley, Extrinsic Frank type (2,3)n and, Multilayer Frank type (4,2), 8H. The optoelectronic study showed that expansion of SFs within the trapezoidal defect is greatly hampered by sessile dislocations and that trapezoidal defects are spread on multiple basal planes; Electron beam induced current (EBIC) imaging showed that dislocations within this defect act as strong sites of carrier recombination which is likely to have an impact on the on-state transfer characteristics of SiC devices. EBIC and Transmission electron microscopy (TEM) revealed that trapezoidal defects come from the substrate and propagate into the epilayer. Furthermore, device electrical measurements showed that as the percentage coverage of trapezoidal defects increases within the active area of a MOSFET device, the on-state resistance increases. Body diode stressing measurements are in agreement with the statement that expansion of SFs within the trapezoidal defect is greatly hampered by sessile dislocations which in part is benecial to body diode degradation. Indeed, the results showed that trapezoidal defects do not degrade devices as much as SFs that can freely expand. Overall, our conclusions nd that trapezoidal defects should still be considered non-killer at low percentage coverage in the active are, but eorts by substrate and epilayer manufacturers need to be made to erase their occurrence. The origin, formation mechanism and behavior of "Star-defect" under device operation were investigated. This study shows that the in-grown star-shaped defect originates in the substrate from on-axis grown boule and propagates until the epilayer/substrate interface. It has a highly strained center core, with primary arms and novel secondary dislocation arrays that were found to be emanating from the primary arms. The secondary arrays are found to be prismatic faults formed from intersection of BPDs. Star-defect has BPDs aggregate present along its core as well as at intersections between primary arms/secondary arrays. These aggregates are nucleation points of SFs expansion that propagate from multiple depths until reaching the epilayer. The total impact on device yield is critical for a single star-defect with secondary arrays as the defect spans a large wafer area (5 by 5 cm-2). A wafer having multiple star-defects will pass initial MOSFETs screening-tests making the devices available on the market but a rapid electrical degradation of the MOSFET is expected. The results presented are crucial to industrial manufacturers in order to assess device reliability. As a result of this work, SiC devices' electrical yield and the potential degradation can be sampled in real time during fabrication.
-
Constituent phases and their stability in a multi-principal element alloy filler for brazing of nickel-base superalloysBrazing is one alternative for joining and repair of nickel-base superalloys, in which only a filler material is melted, and filling of a joint-gap or crack is accomplished via surface wetting and capillary action while the base material remains solid. Design constraints on melting temperature and chemical compatibility with the base material guide the selection of a filler alloy composition. An ideal filler should minimally alter the microstructure of the nickel-base superalloy component at the braze site, particularly avoiding the introduction of potentially detrimental non-native constituent phases. The most common filler alloys in contemporary industrial use employ boron and silicon as melt-point depressants, which often introduce embrittling boride and/or silicide particles, and therefore fall short of this objective. A multi-principal element alloy (MPEA) filler was investigated in this work as a novel alternative. Through a suite of computational strategies considering melting temperature, calculated phases, and interactions with nickel-base superalloy substrates, this MPEA was selected from a composition subspace that is robust in the stability of a single, disordered FCC solid solution. It was therefore hypothesized that use of this filler would render the introduction of detrimental non-native phases to the braze-repair site avoidable. To evaluate this hypothesis, it was important to consider the filler’s tolerance to compositional change driven by dilution (i.e., mixing with the base material) at the immediate point of brazing, as well as the thermal phase stability following both short-term and long-term exposures to high-temperature operating conditions. Synchrotron X-ray diffraction is ideally suited to assess and map constituent phases in a complex brazed microstructure because it offers site-specific structure characterizations at the micron-scale and can detect low volume-fraction constituents. Nevertheless, an MPEA filler as a complex, concentrated alloy exhibits deviations from powder-pattern intensities in as-solidified microstructures and stoichiometric deviations from pure phases, leading to significant challenges in assessing the microconstituents using a conventional Rietveld refinement or strategies based in the powder diffraction file. To overcome these challenges, a novel analytical methodology was developed that synthesizes thermodynamic simulations, existing literature data for crystal structures, and experimental diffraction data coupled with supplemental scanning electron microscopy characterizations. Employing this methodology, experimental investigations were conducted using both Alloy 600 and Alloy 738LC as base materials, to respectively highlight brazing of a solid solution strengthened nickel alloy and a precipitate strengthened nickel alloy with greater industrial relevance. The major constituent phase in either braze was a single, disordered FCC solid solution. The absence of intermetallic phases indicated tolerance of the filler to dilution-driven compositional changes, despite the introduction of seven new elemental species to the filler in the case of the Alloy 738LC braze. The only non-native phases introduced to the as-brazed microstructures of either base material were oxides of Al, Cr, and Mn. The source of oxygen introduction was traced back to the laboratory-scale manufacturing conditions at the original MPEA casting step, indicating that oxide introduction is mitigable by employing more advanced manufacturing processes with better control of atmospheric elements. The only phase evolution detected during short-term high-temperature exposures was the precipitation and growth of additional Cr2O3 oxides, indicating that the MPEA filler possesses short-term thermal phase stability if its dissolved oxygen content can be reduced in manufacturing. Predictive assessments of the long-term phase stability of the MPEA filler using thermodynamic simulations indicated that the filler material is expected to outperform the base material in the long-term suppression of topologically close packed intermetallic phases. Despite the presence of contaminant oxides, the as-brazed ductility of Alloy 600 brazes using the MPEA filler was ten times greater than that offered by a conventional boron-containing filler, with comparable strengths between the two fillers. Variations in total elongation among individual specimens were attributed to variations in the oxide size and distribution. Alloy 738LC brazes using the MPEA filler are also expected to exhibit ductile behavior due to the absence of brittle microconstituents, although the suppression of γ’ is expected to negatively impact strength. A mechanical performance assessment of Alloy 738LC brazes and the re-introduction of γ’ to this microstructure are identified as areas for future work.
-
Development of poly-Si/SiOx passivating contacts for advanced Si photovoltaicsAs the Earth's population grows and nations become more developed the need for energy continues to rise. The climate change brought about by warming temperatures from humanity's greenhouse gas emissions makes it apparent that any future energy sources must be clean and renewable. Of the commonly known renewable energy sources: solar, wind, geothermal, and hydroelectric, the Sun provides by far the largest amount of power to the Earth daily at around ~1000 time more than global power consumption. Because of this incredible potential, solar photovoltaics (PV) are one of the most promising and most studied renewable energy sources. The current solar PV market is dominated by Si solar cells based on the passivated emitter rear cell (PERC) architecture, but newer technologies which can provide higher efficiencies are on the rise. One such technology is the poly-Si/SiOx passivating contact, which provides better passivation and enhanced carrier selectivity compared to PERC contacts. Cells based on the poly-Si/SiOx contacting scheme reach efficiencies >25% in two-sided devices, >26% in the interdigitated back-contact (IBC) architecture, and show promise of reaching high efficiencies in more specialty applications such as long-distance power transfer. We develop the field of poly-Si/SiOx passivating contacts further through studies at atomistic and device levels, using experimental and simulation techniques to solve mechanistic and practical problems. We first study how hydrogen transports from passivating dielectric layers into the passivating contacts to provide passivation. Our results demonstrate that using Al2O3 as a capping layer on SiNx provides enhanced passivation following fast-firing. Next, we study poly-Si/SiOx passivating contacts in the IBC architecture. We see how dopants can contaminate the isolation region between doped fingers to cause shunting. We also demonstrate how this shunting can be prevented by a proposed trap-assisted compensation mechanism. Next, we provide a process by which shunting across this region can be mitigated in complete cells through a self-aligned etching process. Lastly, we demonstrate the use of poly-Si/SiOx passivating contacts in a novel "minicell" architecture designed for high-illumination laser power beaming applications and show that poly-Si/SiOx contacts enable these devices to achieve efficiencies >40% without being limited by series resistance.
-
Modeling the temperature response of small rivers to land cover changes using satellite-based spatial dataThe influence of urbanization and land cover alteration on water quality, including river temperatures, has important ecological implications. However, there is a dearth of information on temperature of rivers smaller than about 60 m wide (approximately fifth order and below), which constitute roughly 97% of total global stream length: for such rivers, field collection of temperature data is labor intensive and often covers short time scales, while satellite-based temperature predictions are often inaccurate for smaller rivers. This lack of high-resolution spatial temperature data has hindered large scale assessment of river temperature patterns both spatially and temporally. This work aims to model the temperature of small, especially urban, rivers across the contiguous United States without requiring field data. To generate sufficiently high-resolution temperature data, a machine learning model was developed for high-accuracy, satellite-based stream temperature estimation, trained with United States Geological Survey stream temperature gage data. The first model developed (called “TempEst”, for Temperature Estimation) had a median validation Root Mean Square Error across gages of about 1.7 K. Building on the outputs from this model as well as validation against paired stream temperature gages, a second machine learning model was developed to predict the relationship between longitudinal changes in river temperatures and nearby land cover conditions, both along the riverbanks and in the general vicinity (without the specific use of satellite-based temperature data). Validated against upstream-downstream pairs of stream temperature gages, the second model (called “LOST”, for Longitudinal Stream Temperatures) had a median downstream Root Mean Square Error across rivers of 1.2 °C. LOST was developed to predict changes in temperature of small rivers anywhere in the contiguous United States based on publicly available land cover, climate, and regional data.
-
Additive manufacturing of refractory alloysRefractory alloys are known for their high temperature capabilities and are intended for use at ultra-high temperatures above even 1200°C, which exceeds the capabilities of conventional superalloys. Additive manufacturing (AM) is an attractive alternative processing route because refractory alloys are difficult to fabricate through traditional methods; AM can form a near net-shape part with a tailored microstructure. This work seeks to evaluate the solidification behavior of refractory alloys under AM conditions and establish corresponding solidification models. Two binary alloys, Mo30Nb and Nb7.5Ta, as well as the commercial Nb alloy, C103, and one refractory high entropy alloy (RHEA), MoNbTaTi, were subjected to single track melts in a laser powder bed fusion (LPBF) machine. Each melt track was evaluated in the scanning electron microscope (SEM), which determined that some AM conditions used led to non-ideal behavior. Electron backscatter diffraction (EBSD) revealed that the Nb7.5Ta, C103, and MoNbTaTi all exhibited new grain nucleation to various extents while the Mo30Nb exhibited primarily epitaxial growth. This was found to be inconsistent with solidification models that were developed for each alloy and each set of AM conditions to predict the columnar to equiaxed transition behavior (CET) of the experimental microstructures. This disparity was largely attributed to the input parameters for the solidification model, which were developed using Thermo-Calc and SYSWELD. These input parameters include the alloy specific Gibbs-Thomson coefficient and thermophysical properties, the solute specific liquidus slopes and partitioning coefficients, as well as the process specific thermal gradients. Each input parameter was evaluated to determine the likely changes required for the modeled values to generate solidification models that better correspond to what was observed experimentally.
-
Design and synthesis of polymeric semiconductors for perovskite applicationsPerovskite solar cells (PSCs) have demonstrated remarkable efficiency growth in their brief history, and are considered to be of exceptional potential for commercialization due to their excellent absorption properties, varied deposition methods, and compatibility with other solar technologies. Furthermore, perovskites are demonstrating high potential for other applications, such as perovskite light emitting diodes (Pero-LEDs), lasers, and radiation detectors. Most perovskite based devices use hole transporting materials (HTMs) to assist in charge separation and current generation. The three main categories of HTM are inorganic materials, small organic molecules, and polymeric materials. Organic materials typically provide the highest efficiencies for these devices, but have several drawbacks including low economic viability, lack of flexibility for use with the various perovskite absorber layers (PALs), and difficulty of application in the multiple device architectures that exist for these devices. This dissertation primarily describes the design and synthesis of new polymeric materials to improve the processibility and interfacial interactions of HTM and PAL, leading to high efficiency, high stability, and low cost PSCs. Our current research into HTMs takes a four-pronged approach; We found that utilizing the Buchwald-Hartwig amination protocol using primary aryl amines and aryl dihalides afforded highly reproducible, high yielding family of polymers, which could be purified by a simple sequence of precipitations. Appropriate selection of pendant functional groups, such as electron donating methoxy, or electron withdrawing fluorine, allowed for highest occupied molecular orbital (HOMO) tuning, as did the utilization of electron rich carbazole versus the neutral fluorene in the polymer backbone. Control of the glass transition temperature (Tg), a characteristic vital to extended lifetime at elevated temperature, was demonstrated by manipulation of the alkyl side chains in the polymer, which allowed for a balance of solubility and improved Tg. Finally, side chain engineering of the polymers, incorporating more hydrophilic functional groups, was explored to improve the processibility of the PAL on top of the polymer HTMs. This allowed for the manufacture of devices that did not require an interfacial layer or UV/Ozone treatment to form a consistent perovskite film, removing a variable in the device, as well as reducing processing time and cost
-
Characterization of high deposition rate wire arc directed energy deposition of 316L stainless steel for pressure boundary componentsWire-arc directed energy deposition (WA-DED) or wire-arc additive manufacturing (WAAM) presents a novel fabrication method for construction of pressure retaining components within nuclear power plants; reducing lead times for construction of replacement parts and potentially preventing millions of dollars in losses from power plant down time. However, ASME code changes are required before WA-DED 316L can be utilized for this application. This work aims to support these code changes. A WA-DED process parameter study was undertaken to better understand how travel speed, interpass temperature, and filler wire metal selection impact the microstructure and mechanical properties of 316L. It was found that interpass temperature and travel speed displayed no significant impact on the mechanical properties of WA-DED 316L within the parameter ranges evaluated. However, the selection of 316LSi filler wire metal resulted in an increase in yield strength, tensile strength, and ductility over 316L filler metal, likely due to solid solution strengthening and a reduction in stacking fault energy. Across conditions, heat treated samples exceeded the ASME minimums of a yield strength of 172 MPa, a tensile strength of 482 MPa, and an elongation of 43.6% for 316L at room temperature. A heat transfer model was developed to allow for prediction of the thermal history of the WA-DED process. The heat transfer model was used in conjunction with solidification models to make predictions of dendrite spacing and growth morphology showing good agreement with experimental measurements. Further work was undertaken to understand the role of hatch spacing on lack of fusion porosity in WA-DED builds. By increasing the ratio of hatch spacing to weld bead width, lack of fusion defects were observed in-situ using an infrared thermal camera. It was found that lack of fusion begins to form at a critical hatch spacing to bead width ratio of over 0.68. Lastly, a large-scale WA-DED 316L body was produced to gain understanding of print strategy development and part design for WA-DED. This large-scale body was separated into three distinct sections for production and surfaces were overbuilt by approximately 6.4 mm to allow for post process machining. A 323 mm (12.7 in) tall 316L representative valve body was successfully produced through WA-DED following the print strategy developed.
-
New lens on art history: using complex network analysis and unsupervised machine learning, AVisual art is a wholly complex and inherently human creation not easily analyzed and interpreted by digital technology. The subjectivity of art makes its interpretative understanding elusive to machines while virtually instantaneous to the human viewer. This project attempts to demonstrate that there is a relationship between the measurable visual characteristics and the communicative characteristics of art. In doing so, we hope to offer a machine-based software tool that supplements the traditional critical approach to historical art found in art history and art theory. This AI-generated perspective will offer innovative insights to the inherent interpretative information found in art. This project's methods are seated in the creation of a python-based feature extraction software. The software is an analog to the pluralistic critical approach of art theory. It abstracts images of historical paintings into complex network representations that contain the digital equivalent of formalist elements and principles of design present therein. By measuring the images' network representations, we obtain quantitative descriptions of their innate visual features. We, then, reduce the dimensionality of the measurement data set and find a clustering of the images. From those clusters, we can draw mathematical conclusions about the interpretative characteristics of the images held within. We postulate that the evaluative conclusions enabled by our method's AI-generated art movements will reach beyond those present in traditional art theory. We measure the interpretative precision of the clustering we obtain using the precision and recall performance measures. We compare the resulting performance from our software to that of a random clustering of images. In doing so, we prove that our software indeed performs better and is statistically distinct from a random grouping of paintings in terms of critical and formalist evaluation. Beyond that, we show that our resulting clustering has greater success in terms of the performance metrics than the critically accepted historical art movements. These results show that, using complex networks to embody formalist elements and principles of design, measuring those networks, and clustering the paintings based on that data, we are not only able to create distinct groupings of images with common formalist components but common critical interpretations as well. Because of this, we can confirm the existence of an untapped empirical relationship between machine-measurable visual characteristics of images and the communicative concepts held within those images. That we obtained performance better than the historical movements shows that our methods offer the first steps to discerning and building on this correlation.
-
Wavefield reconstruction using seismic interferometry and compressive sensingSeismic data acquisition and processing are essential steps for seismic exploration, the determination of deep earth structure, and subsurface monitoring. Common problems in seismic data acquisition are missing or unusable receivers, and a low signal-to-noise ratio of recorded signals. These problems apply to seismic interferometry, a passive-source technique to estimate inter-receiver wavefields. In this thesis, I propose novel methods that use seismic interferometry and compressive sensing to mitigate these problems. I use inter-source interferometry to estimate body waves that propagate between two earthquakes from waves that are recorded at a surface receiver array. Inter-source interferometry, which turns an earthquake into a virtual receiver, allows receivers to be virtually placed at earthquake locations (e.g. inside volcanoes and subduction zones) that are normally inaccessible for receiver installation. To accurately estimate inter- source body wavefields, I derive a criterion for how densely receivers must be spaced to retrieve the waves that propagate between earthquakes. Using waves recorded near the San Andreas Fault, I estimate the inter-earthquake wavefields. The problem of missing or unusable receivers is important for receiver arrays that do not regularly sample the wavefield at the surface. I propose a compressive-sensing-based multi-source wavefield reconstruction to alleviate the problem of missing or unusable receivers. Using the Fourier and Curvelet domains for sparse transforms, I show that a multi-source method, which reconstructs correlated wavefields at the locations of unavailable seismometers from correlograms of all available virtual sources, improves wavefield recovery compared to single-source reconstruction, which uses a correlogram from a single virtual source. I show successful applications of multi-source reconstruction for wavefield-recovery improvement over single-source reconstruction using synthetic data on linear and areal arrays and present a data example using distributed acoustic sensing data. The methodology is, in principle, applicable to active source seismic surveys. Also, I develop weighted compressive sensing to mitigate the imprint of noise on the reconstructed wavefield, and to impose a priori information about the nature of the recovered wavefields. Wavefields obtained from seismic interferometry may contain spurious arrivals. Using the Fourier basis as a sparse transform, I show that weighted compressive sensing can suppress these spurious arrivals.
-
Surface modification of nanomaterials for electronically tunable polymer compositesDeveloping a sustainable energy future has recently come to the forefront of many research endeavors in academia, industry, and government laboratories. One approach that has been presented as a viable, long-term option is the conversion of solar radiation to power, using perovskite solar cells. These devices have garnered a great deal of attention because of their ability to absorb large amounts of light, superior charge-carrier mobility, scalability, and high efficiencies. For these devices to be produced on an industrial scale, further optimization, and development of layers such as the hole transport layer (HTM) will require significant advancements. Current commercially used HTM are polymers such as poly[bis(4-phenyl)(2,4,6-trimethylphenyl)amine] (PTAA) and poly(3-hexylthiophene-2,5-diyl) (P3HT), which require multistep syntheses with complex purifications leading to high cost materials. Until recently, these were the challenges that researchers have been attempting to address through new synthetic approaches. Fortunately, a library of new triarylamine based polymers with tunable properties have been developed, addressing many of these constraints while producing efficiencies and stabilities exceeding PTAA. Although based on the hydrophobic nature of these polymers, they do have desired interfacial interactions with perovskite ink during the solution deposition process. This unfavorable condition often requires the addition of an interfacial wetting layer which costs upwards of $2,000/g. This thesis describes the enhancement of wetting properties within triarylamine based polymers, while retaining their ability to behave as excellent HTMs. The approach was to maintain the backbone of these polymers while adjusting the long alkyl chain moieties to increase hydrophilicity, ensuring that the optical and electronic properties would not be affected. By introducing methoxyethoxy(ethane) chains, the inherent affinity for polar solvents increased significantly. With this in mind, small equivalence of these wettable monomers were introduced into the polymerization resulting in random terpolymers with improved wettability. Different incorporation ratios were explored for both carbazole and fluorene-based polymers, resulting in HTMs with enhanced perovskite ink affinity that did not require additional interfacial wetting layers. The second portion of this thesis focuses on tuning the work function and conductivity of these wettable polymers through functionalized nanomaterial doping. By surface functionalizing nanoparticles with electron withdrawing/donating ligands, the work function and conductivity of those particles can be adjusted. The approach focused on three main nanomaterials: graphene, carbon nanodiamonds and MXenes. Synthetic routes were developed for the surface functionalization of each material, followed by extensive characterization and eventual doping into the wettable polymer termed CzMee10. Doing so allowed for tunability of the polymer/nanomaterial composite’s work function over a range of 0.51 eV depending on the nanomaterial and ligand used. The conductivity at the time of writing this thesis is still being evaluated. This analysis and characterization encompassed much of the final part of the thesis and is followed by a short introduction to future projects as well as a recap of the interdisciplinary skills utilized to achieve these advancements.
-
Towards a mechanistic understanding of contaminant attenuation and greenhouse gas emissions in open-water engineered wetlandsEngineered wetlands offer a sustainable supplement to conventional water and wastewater treatment by harnessing biological processes which occur naturally in the environment. In macrophyte-free open-water engineered wetlands, design parameters of a shallow water column and geotextile liner select for a benthic microbial biomat with parallels to periphyton biofilms in shallow streams. This dissertation aims to disentangle the microbial interactions that govern contaminant biotransformations and greenhouse gas emissions within this benthic biomat community, leveraging a demonstration-scale open-water engineered wetland in Corona, California. Through an integration of field-scale genome-resolved metatranscriptomics, porewater profiling, and greenhouse gas fluxes with inhibition microcosms manipulating redox conditions to interrogate specific metabolisms, pathways which occurred simultaneously in-situ were decoupled and associated with contaminant transformations. First, photosynthesis, nitrification, and denitrification were associated with the biotransformation of a suite of pharmaceutical compounds, including novel linkages of nitrate and nitrous oxide reducing activity with the biotransformation of an anti-viral (emtricitabine) and antibiotic (trimethoprim). Next, methane-oxidizing activity, catalyzed by the particulate methane monooxygenase as confirmed by field metatranscriptomics and inhibition microcosms, stimulated the biotransformation of sulfamethoxazole, an antibiotic that was highly recalcitrant under all other surveyed conditions. Finally, mechanistic insights were synthesized to construct a series of models which estimate the contributions of benthic metabolisms to greenhouse gas emissions, identifying methane-oxidizing bacteria as important ecological filters of climate forcing. Taken together, these findings are discussed in the context of open-water engineered wetland management and design, environmental contaminant fate and transport, and the critical intersection of these themes with climate change.
