• Organic semiconductor design for increased performance and stability of perovskite solar cells

      Sellinger, Alan; Schloemer, Tracy H.; Knauss, Daniel M.; Domaille, Dylan; Rockett, A. (Angus) (Colorado School of Mines. Arthur Lakes Library, 2019)
      Perovskite solar cells (PSCs) have received considerable attention due to the absorbing layer’s excellent semiconducting properties and simple deposition methods that allow for impressive device-level efficiencies at potentially much lower cost than state-of-the-art technologies. Most PSCs use organic hole- and electron-transport materials for highly efficient devices. While many organic charge-transport materials used allow for high device-level efficiencies, they often are the bottleneck with regard to thermal stability in the context of the materials in the entire device stack. This dissertation primarily describes the design, synthesis, purification, and characterization of new charge-transport materials and dopants to develop molecular structure and performance relationships for application in highly durable PSCs. Our current hole-transport layer (HTL) research focuses on developing both hole-transport materials and dopants. HTL design is one method with a high return on investment for PSC stability and lifetime. We found radical triarylamine salts as dopants must contain two electron-donating groups para to the triarylamine to prevent dimerization that renders the dopant ineffective. A stable dopant combined with a triarylamine-based, high glass-transition temperature HTM allows for improved device-level stability at elevated temperatures (50 C). To extend this system to further improve device-level properties, HTMs with thiomethyl substituents for perovskite active layer (PAL) passivation and cross-linkable HTMs were synthesized and characterized with the aim to improve device-level efficiency and operational stability at elevated temperatures respectively. Upon device-level optimization, these routes were not ultimately achieved due to unexpected fabrication considerations. With regard to electron-transport layers (ETLs), naphthalene diimide-based (NDI) materials have recently emerged as a potential competitor for C60, the state-of-the-art organic ETL due to their impressive morphological stability and intrinsic electron mobilities. Due to poor solubility in solvents orthogonal to the perovskite active layer, NDIs are not compatible with solution-processing. A library of NDI-based ETLs was synthesized and characterized for improved solubility and frontier orbital energetic alignment with a variety of PALs. Optoelectronic properties were characterized which show successful frontier molecular orbital energetic tuning. Structural motifs for moderately improved solubility in desired solvents were also identified. The PAL contains a number of species that can participate in a variety of chemical reactions (e.g., Brønsted acidic organic species like methylamine, redox active halides, etc.). The organic charge-transport layers must survive these conditions during device operation. Little work has been done to develop design criteria for charge-transport materials based upon compatibility with the PAL. Towards this end, perovskite precursor ink impurities were identified and characterized. Acid-base reactions between aliphatic amine-lead coordination complexes and free aliphatic amines produce an aliphatic ammonium and lead-amide species. These findings highlight the importance of perovskite precursor ink chemistry and potential impurities in the thin film after deposition.
    • Theoretical and experimental approaches to elucidating the redox-driven degradation mechanisms of perfluoroalkyl substances

      Vyas, Shubham; Van Hoomissen, Daniel; Wu, David T.; Higgins, Christopher P.; Sellinger, Alan (Colorado School of Mines. Arthur Lakes Library, 2019)
      Chemical scientists catalyze technological innovations across numerous disciplines in modern society. To some environmentalists, however, it’s evident that our inventions can do more harm than good. A lack of hindsight and foresight continually plague human progress as ongoing chemical pollution accompanies our technological discoveries and destroys the biosphere. Our deficiency in these aspects is consequentially related to our poor understanding of the fundamental physiochemical properties of recalcitrant compounds. This continues to stymie our engineering efforts to mitigate the harm of environmental pollution. To fill these knowledge gaps, my research focused on elucidating and contextualizing the chemical mechanisms associated with the remediation and destruction of persistent organic pollutants. Perfluoroalkyl substances (PFASs), anthropogenic contaminants of emerging concern, will be the subject of a large fraction of this work and are extensively described from a physical-organic perspective. A fundamental premise of this work the application of complementary experimental, computational and spectroscopic techniques can facilitate a bottom-up molecular approach to tackle these macroscopic existential problems. PFASs are a class of compounds which have one or more perfluorinated alkyl chain(s) attached to a diverse and extensive set of organic functional groups. Because of the unique properties of C—F bonds, these compounds are utilized in numerous commercial, industrial and pharmaceutical applications, and their continued use in these aspects constitutes billions of dollars in economic value. Their economic and technological usefulness is offset by their inability to biodegrade in the environment, and subsequently, biosphere wide contamination of fluorinated compounds has emerged as a worldwide dilemma. Much has been done to quantify the risk of PFASs and understand the degree to which they are released into the environment. Recently, bench-top investigations utilizing photochemical techniques have shown promise to degrade PFASs; however, to date, degradation-based methods are not implemented on larger scales. To this end, this thesis is focused on a synergistic computational-experimental approach to understand the degradation mechanisms of fluorinated organic contaminants, especially those associated with UV-mediated redox chemistry. First, many fundamental properties of PFASs are described, from their synthesis, to their use and destruction. Next, I will describe our approach to predict the ability for Co(I)-catalysts to reductively cleave C—F bonds in PFASs. This is followed by developing an understanding about their redox behavior and the subsequent isomerization reactions of activated degradation intermediates. Inspired by our work on isomerization events in fluorinated radicals, parallel methods were also used to describe the relative stability of other environmentally relevant radical species, allyl and benzyl radicals. Density functional theory (DFT) was imperative to understand the vital attributes for each component of this thesis and was essential in the development of structure activity relationships for a broad suite of PFASs. In short, the polar functional group and the chain length of the perfluoroalkyl acids was found to govern which C—F bond would cleave under reductive conditions. In conjunction with DFT, ongoing development of laser flash photolysis techniques will support the measurement of the reduction rate constants between PFAS and UV-generated hydrated electrons and complements ongoing work to compute the rate of electron transfer with Marcus theory. These methods will help us improve our understanding of existing practices for the remediation of PFASs and could facilitate the development of novel technologies to remediate a wide-range of PFASs-impacted natural waters.
    • Development of thermal field-flow fractionation for the characterization of hybrid nanoparticles and polymers with complex architecture

      Williams, S. Kim R.; Smith, William Conner; Wu, Ning; Wu, David T.; Trewyn, Brian (Colorado School of Mines. Arthur Lakes Library, 2019)
      Field-flow fractionation (FFF) is a family of analytical techniques designed for the separation of macromolecules and colloids on the nanometer to micron scale. The well-established theory governing FFF separations permits physiochemical properties to be calculated from measured analyte retention times. This allows FFF to be leveraged for not only separation of nanoparticles and polymers but also as a characterization tool to obtain physiochemical distributions. Thermal FFF is unique as the separation mechanism is based on the differential transport of analytes in the presence of a temperature gradient. This thermophoresis (or thermal diffusion - DT) is dependent on interactions at the analyte-solvent interface yielding transport driven by compositional differences. This permits the development of novel approaches to separating and characterizing complex polymers, functionalized, and hybrid nanoparticle systems. An in-depth assessment of competing thermophoretic theories was essential to identifying potential key levers (experimental conditions) that could manipulated to control ThFFF retention. This knowledge was leveraged to develop the unique analytical capabilities of ThFFF. First these principles were employed to separate and determine the compositional distributions of inorganic metal-metal oxide hybrids such as multi-lobed Pt-Fe3O4 nanoparticles. Results indicate that correlation of DT to both particle surface and bulk properties maybe dependent on the dielectric nature of the solvent. Subsequently, the role of solvent quality and presence of additives (salts, surfactants, ligands etc.) on polymeric particles and polymer stabilized metallic particles was studied. Results suggest that particle DT can be tuned via modification of a particle surface through the use of adsorbates. The affinity of the adsorbates for specific faceting on the analyte surface could be further exploited to enhance separations by morphology. Investigation of analyte-solvent interactions and the role of solvated conformations of organic-inorganic hybrids was strongly informed by studies of soluble polymer systems. The thermophoretic behavior of hyperbranched polyesters with various degrees of branching was examined showing that the presence of linear, cyclic, and branched polymers could be identified and polymer architecture distributions could be determined using ThFFF first principles. In conclusion, Thermal FFF has proven to be a unique technique capable of separating and characterizing complex colloidal solutions.
    • Mineralogy and petrogenesis of the California Blue mine aquamarine- and topaz-bearing pegmatite deposit, San Bernardino County, California

      Gysi, Alexander; Pauly, Carolyn; Pfaff, Katharina; Wendlandt, Richard F. (Colorado School of Mines. Arthur Lakes Library, 2019)
      The California Blue Mine Pegmatite is a recently discovered gem aquamarine- and topaz- bearing deposit near Yucca valley in southern California, and is the first significant gem-bearing pegmatite discovered north of the Southern California gem districts (Hunerlach, 2012). Aside from large-scale mapping by the USGS and field work necessary for mining, little research has been conducted on the deposit. The purpose of this thesis is to provide a catalogue of the mineralogy, textures, zoning, and geochemistry of the California Blue Mine pegmatite, and to synthesize these observations into a petrogenetic model. This research combined mineralogical, textural, and geochemical field and hand sample observations with lab analyses, including optical microscopy, field emission scanning electron microscopy (FE-SEM), automated mineralogy, electron microprobe analysis (EMPA), and laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS). A distinct textural and mineralogical zoning has been identified in the pegmatite, including a border zone, wall zone, intermediate zone, and core zone, as well as several miarolitic pockets. The presence of these zones and pockets indicate a transition from magmatic to hydrothermal processes during pegmatite formation, an assertion which is supported by geochemical and textural data in beryl. Beryl shows three types of textures coinciding with three stages of crystallization: Type i) distinct to gradational concentric core zoning, produced in the intermediate zone by magmatic processes; Type ii) transitional patchy rim zoning, produced in the core zone by magmatic processes and mild metasomatism, and Type iii) pervasive alteration rims caused by intense hydrothermal alteration. Beryl cores are enriched in Al, and rims are enriched in Sc, Na, Fe, and Cs; the coupled replacement of Sc3+ and Fe2+ for Al3+ on the octahedral site with Na+ and Cs+ on the beryl channel site is the most important substitution mechanism in the beryl and defines the transition from Type i) to Type iii) textures. A combination of this octahedral site substitution and coupled substitution of Li+ for Be2+ on the Be tetrahedral site with alkali cations in the channel site characterizes the Type ii) and Type iii) textures. Metasomatic rims of beryl are characterized by enrichment in Li and Cs. The overall evolution of the pegmatite can be subdivided into three stages based on mineral paragenesis and beryl geochemistry. Stage 1 represents crystallization of the quartz, perthitic K-feldspar, albite, biotite, spessartine, and Type i) Al-enriched beryl in the border, wall, and intermediate zones of the pegmatite, produced by fractional crystallization and subsequent melt enrichment in Li, Be, and Sc. Stage II includes the formation of graphically intergrown quartz and albite and Sc-, Na-, and Fe-enriched Type ii) beryl in the pegmatite core by metasomatism from an aqueous fluid separated from the original melt, and Stage III represents the final crystallization of the miarolitic pockets from ponding of the aqueous fluids, producing gem smoky quartz, topaz, and Type iii) beryl enriched in Li and Cs. We conclude that the California Blue Mine pegmatite contains mineralogy similar to NYF-type pegmatites but metasomatic Li and Cs overprinting similar to LCT-type pegmatites, and thus requires more research to classify as either LCT- or NYF-type.
    • Interpreting sources and endocrine active components of trace organic contaminant mixtures in Minnesota lakes

      Higgins, Christopher P.; Guyader, Meaghan E.; Bellona, Christopher; McCray, John E.; Furlong, Edward T.; Ranville, James F. (Colorado School of Mines. Arthur Lakes Library, 2019)
      On-site wastewater treatment systems (OWTSs) are a suspected source of widespread trace organic contaminant (TOrC) occurrence in Minnesota lakes. TOrCs are a diverse set of synthetic and natural chemicals regularly used as cleaning agents, personal care products, medicinal substances, herbicides and pesticides, and foods or flavorings. Wastewater streams are known to concentrate TOrC discharges to the environment, particularly accumulating these chemicals at outfalls from centralized wastewater treatment plants. Fish inhabiting these effluent dominated environments are also known to display intersex qualities. Concurrent evidence of this phenomenon, known as endocrine disruption, in Minnesota lake fish drives hypotheses that OWTSs, the primary form of wastewater treatment in shoreline residences, may contribute to TOrC occurrence and the endocrine activity in these water bodies. The causative agents specific to fish in this region remain poorly understood. The objective of this dissertation was to investigate OWTSs as sources of TOrCs in Minnesota lakes, and TOrCs as potential causative agents for endocrine disruption in resident fish. Three research efforts were executed to investigate these topics: examining chemical and biological signatures of OWTS proximity in Minnesota lakes (Chapter 2), prioritizing potential causative agents of endocrine disruption using liquid chromatography tandem mass spectrometry (LC-HRMS) (Chapter 3), and conducting a suspect search of OWTS-associated LC-HRMS features at an adjacent Minnesota lake (Chapter 4). In Chapter 2, traditional targeted aqueous analyses indicated higher concentrations of TOrCs at locations more proximal to residences with OWTSs. Residential proximity also corresponded to feminization of male sunfish. The particular contaminants detected at these locations are considered weak indicators of wastewater presence in environmental compartments and inactive with current metrics for endocrine activity. Unexpected features unique to sites with pronounced endocrine disruption were identified in Chapter 3; however, these components were still considered endocrine inactive by current toxicology metrics. This suggests that temporal resolution in sampling was too low for this chronic-toxicity endpoint, or current regulatory efforts underestimate the effects of these contaminants as environmental mixtures. Finally, Chapter 5 indicated that LC-HRMS analysis and passive sampling allowed for identification of a broader suite of OWTS-associated compounds in adjacent lake water, but that higher spatial resolution was required to refine lake-specific OWTS-compound interest lists. The results from this dissertation encourage further investigation of residential inputs of TOrCs to Minnesota lakes, particularly prioritizing locations with endocrine disruption, so that local regulatory agencies may effectively manage these highly valued state resources.
    • Developing and identifying physically based Li-ion battery models to inform real-time control applications

      Kee, R. J.; Vincent, Tyrone; Weddle, Peter J.; DeCaluwe, Steven C.; Jackson, Gregory; Mohagheghi, Salman (Colorado School of Mines. Arthur Lakes Library, 2019)
      The enclosed thesis develops and identifies physically based battery models. The physically based battery models bridge fundamental chemical, electrochemical, and transport scales to system-level applications. The fundamental physics captured in such detailed models provide qualitative and quantitative insight for effective battery design and implementation. Because fundamental models are too computationally expensive to run at scale, systematic reduction techniques are introduced to identify low-order models that can be run in real time. Many techniques are implemented throughout the thesis to inform battery/pack level models of fundamental electrochemical physics. The techniques presented not only upscale important battery dynamics, but also can be used for practical implementation in battery management systems. The thesis is generally formatted to (1) introduce and derive common approaches to modeling important battery physics at the porous media scale (i.e., transport/kinetics between current collectors), (2) develop upscaling techniques to inform full-cell battery models of underlying microscale (electrode) composition and mesoscale (current collector wrapping) architecture, (3) formulate a systematic procedure for identifying reduced, low-order, linear state-space models that accurately describe full-cell battery dynamics, (4) implement the identified state-space models into model-predictive control algorithms, and (5) introduce and analyze phase-transformation physics in such a way as to identify important electrochemical impedance realizations. This research has at least four significant contributions to the field. The first novel contribution is developing upscaling techniques that inform full-cell models of mico- and meso-scale compositions. The physically informed full-cell models indicate that the influence of heterogeneous upscaled parameters are most significant during abuse scenarios. The second novel contribution is a systematic procedure to extract linear-state space models from complex battery models. Because the extraction procedure treats the physically informed battery model as a black-box, the same procedure can be extended to real batteries. The third novel contribution are pioneering model-predictive control algorithms that implement the extracted state-space models and respect both measurable and unmeasurable constraints. The final novel contribution of the present thesis is to interpret phase-transformation physics at the electrode particle ensemble scale. The fundamental analysis indicates that phase-transformation electrodes will exhibit history-dependent state-of charge and subsequently a history-dependent electrochemical impedance spectra. Phase-transformation electrode charge history-dependent impedance is validated with experimental results.
    • Fission fragment distribution measurements with time of flight - energy (ToF-E) spectrometers

      Greife, Uwe; Tovesson, Fredrik; Moore, William Phillips; King, Jeffrey C.; Sarazin, Frederic; Gray, Frederick; Wiencke, Lawrence (Colorado School of Mines. Arthur Lakes Library, 2019)
      Neutron induced fission has been studied for over seventy years, however, there is still much to learn about the process. Evidence has been shown for energy dependence of the neutron inducing fission on the fission product yield (FPY) distribution, and there is very little existing data for the FPY of various isotopes of interest as a function of incident neutron energy. The CSM FPY time of flight - energy (ToF-E) spectrometer was developed and measurements were performed with 252Cf(sf) in support of the Spectrometer for Ion Detection in Fission Research (SPIDER) collaboration. A variety of ionization chamber (IC) entrance windows were investigated. Stretched polypropylene and a commercial option from LUXEL were found to be adequate entrance windows. Silicon nitride windows might offer the least amount of energy loss from fission products through the entrance window, however, there is a significant amount of scattering of fission products off of the frame and support structures, which has shown to impact FPY results in the symmetric fission region as well as the upper and lower ends of the FPY distribution. The impact from scattering on FPY results are greatly reduced with either the stretched polypropylene window or the LUXEL window. Single arm operation of the CSM FPY ToF-E spectrometer from 252Cf(sf) was shown to have a potential lower limit on mass resolution of 1.5 amu full width at half maximum (FWHM) for light fission products and 3.0 amu FWHM for heavy fission products, operating with a 40 cm arm length, a stretched polypropylene entrance window (92 μg/cm2 thickness), and 30 μg/cm2 thick carbon conversion foils. Several methods for extracting nuclear charge information from the Frisch grid (FG) signal of the IC were investigated. Nuclear charge sensitivity was exhibited with a potential resolution around 4 amu FWHM. Single arm analysis of SPIDER data from 235U(nth,f) with 70.0 & 67.3 cm arm lengths and 200 nm thick silicon nitride entrance windows showed a potential lower limit on mass resolution of 1.5 amu FWHM for light products and 3.0 amu FWHM for heavy products. Dual arm analysis with the double energy - double velocity (2E-2v) method revealed improvement in the mass resolution showing a potential lower limit possibly better than 1.0 amu FWHM for light products and 1.5 amu FWHM for heavy products. Single arm analysis of SPIDER data from 233U(nth,f) with the same operating conditions as the 235U(nth,f) run showed a potential lower limit on mass resolution of 2-3 amu FWHM for light products and 4-6 amu FWHM for heavy products. Again, using the 2E-2v analysis method resulted in improvement in the mass resolution achieving a potential lower limit of 1.5 amu FWHM for light products and 3.0 amu FWHM for heavy products. Further investigation into the 233U target revealed potential “crud” in the chemical com- position deposited during the production process. This “crud” caused degradation in mass resolution by a factor of almost two, indicating FPY mass resolution dependence on target quality. The goal of unit mass resolution from SPIDER has been shown to be achievable with use of clean targets, 70 & 67.3 cm arm lengths, 20 μg/cm2 thickness carbon foils, and 200 nm thick silicon nitride entrance windows.
    • Modeling spatial geotechnical parameter uncertainty and quantitative tunneling risks

      Mooney, Michael A.; Grasmick, Jacob G.; Hedayat, Ahmadreza; Walton, Gabriel; Zhou, Wendy (Colorado School of Mines. Arthur Lakes Library, 2019)
      In underground construction and tunneling (UCT) works, geological and geotechnical uncertainty is often the most significant source of risk associated with a project. Unforeseen adverse geological/geotechnical conditions can lead to significant construction issues causing reduced tunnel advance rates and schedule delays, cost increases, damage to existing infrastructure, and/or damage to construction equipment. However, current practice falls short of reliably assessing spatial uncertainty in geological conditions and geotechnical parameters. This thesis addresses the spatial uncertainty encountered in UCT works, utilizing the site investigation data from several tunneling projects. The analysis of typical geotechnical site investigation data for tunneling projects demonstrates that there is often sufficient data to assess the spatial correlation structure, and that this structure can vary by engineering soil unit type. Therefore, an approach to modeling the spatial variability and uncertainty of geotechnical parameters, while jointly considering the geological variability and uncertainty, was developed. In addition, the levels of uncertainty in engineering soil unit classification and a representative geotechnical parameter, N1(60), encountered in three distinct UCT projects is assessed. These projects enable a broad assessment of different project scales, degrees of site investigation effort and geological conditions, and the corresponding uncertainties. The results of this analysis reveal that spatial uncertainty can differ significantly both within a project and across multiple projects. Furthermore, while a higher degree of site investigation effort leads to lower average uncertainty in the engineering soil unit estimation, the same is not necessarily the case for N1(60). The analysis also conveys that quantified uncertainty in geotechnical parameters is influenced by other factors in addition to site investigation sample spacing including uncertainty in geological variability and measurement error. This thesis also presents a novel application of geostatistical simulation for risk assessment and mitigation planning for two frequent risks encountered during mechanized tunneling: clogging and cutter tool wear. The quantification of spatial uncertainty in geotechnical parameters enhances the risk assessment as the uncertainty in risk estimates can also be quantified spatially. The results of the proposed mitigation plans (water injection and cutter tool replacements) from the geostatistical model approach were found to be in strong agreement with the project data, validating both the methodology and the value of performing such detailed analysis for tunneling risks.
    • Field development recommendation utilizing 3D geomechanical model for an unconventional play in Colombia

      Tutuncu, Azra; Miskimins, Jennifer L.; Arias Medina, Leonardo; Fleckenstein, William W.; Sonnenberg, Stephen A. (Colorado School of Mines. Arthur Lakes Library, 2019)
      A methodology for a 3D geomechanical model for shale plays based on pre-stack time migration (PSTM), 1D geomechanical model and geochemical analysis are presented in this study to understand an unconventional reservoir in Colombia. Vertical Transversely Isotropic (VTI) was used to calculate the anisotropic geomechanical properties of the shale formation. Exceptional shale plays in Colombia have been encountered through a drilling campaign in the last few years. Geomechanical and geochemical characterization of these shale formations have been used to provide the required information for designing optimal well trajectories and understanding reservoir and completion qualities. Additionally, extensive tri-axial, and Rock-Eval measurements have been included in the study to calibrate the mechanical properties and organic composition from well logs. The critically stressed fracture concept has been included in the geomechanical model to determine the conductivity of the natural fractures in the shale play. Analyses of the in-situ principal stresses, geomechanical properties, mud weight windows, optimal well trajectories, kerogen type and thermal maturity of the formation have been carried out. Hampson-Russell model was used for inversion of the seismic from pre-stack time migration to deliver a 3D volume of P-wave impedance, S-wave impedance and bulk density to have lateral and vertical spatial coverage. 1D and 3D geomechanical models were developed applying commercial and proprietary software packages with customized analysis of the fractured reservoir stress state in the shale play. The pore pressure, litho-facies, geomechanical properties, and Total Organic Carbon (TOC) in specific areas of the reservoir were determined using electrical and image logs, core samples, drill cuttings, and field data from two wells. GOHFER 3D™ software package was used for hydraulic fracturing simulation of selected intervals in Well 1 to identify the types of hydrocarbons in the potential shale plays and to understand the performance of the well for stimulation, regardless of not expecting commercial production rates in this vertical well. A Discrete Fracture Network (DFN) was built for the G horizon in the studied field to predict natural fractures away from the two wells in the field using image logs and seismic attributes. The results were used in recommending specific development scenarios in the shale play.
    • Fundamental electrochemical study on neodymium molten salt electrolysis in fluoride bath

      Taylor, Patrick R.; Liu, Fangyu; Eggert, Roderick G.; Anderson, Corby G.; Seetharaman, Sridhar; Earlam, Matthew (Colorado School of Mines. Arthur Lakes Library, 2019)
      In the recent decades, the clean energy economy has been driving a rapidly increasing demand for rare earth materials with their applications in essential high technologies such as electric vehicles and wind turbines. The technologies for the winning of rare earth metals are developing. As the predominant winning technique, electrolysis of rare earth oxides in molten fluoride systems has been faced with two major problems: one is low energy efficiency and the other is high emissions of perfluorocarbons (PFCs). Therefore, there exists a need for the metallurgy community to address those problems both theoretically and practically, and to develop improved processes for reducing rare earth oxides. This thesis uses the example of neodymium winning to elucidate the fundamental electrochemical properties of the molten fluoride electrolytes and the mechanism of the electrolysis process, and provides a guide to economically win rare earth metals with elevated energy efficiency and decreased emissions for the process. As for the property investigation, this research carried out measurements to determine the liquidus temperatures of the NdF3-PrF3-LiF ternary salt system, the solubilities of Nd2O3 in the electrolytes, and the electrical conductivity of the NdF3-LiF salt. A conductance cell system was developed to investigate electrical conductivity and produce reliable data. The experimental results indicate that the electrical conductivities of the molten NdF3-LiF system between 70 wt% to 85 wt% NdF3 within the range of temperature from 950 °C to 1050 °C range from 4.38 ohm-1 cm-1 to 6.08 ohm-1 cm-1. Furthermore, an empirical equation to estimate the value of the electrical conductivity for a specific molten salt is proposed. A mathematical model regarding the voltage change against current in the molten Nd2O3-NdF3-LiF electrolysis is proposed based on the thermodynamics and kinetics study and validated through experiments and literature observations. The model and the experimental results illustrate that the limiting current of the electrolysis cell increases with the increase of anode surface area, higher energy efficiency can be achieved with reduced electrode distance and more effectively, reduce the submerged depth ratio of anode to cathode. An effective technique to prevent the generation of PFCs is to design the cell conditions which allow the limiting current to be smaller than the critical current.
    • Development of an instrument for the chemical and isotopic analysis of sub-nanomole quantities of gas from individual fluid inclusions, microfossils and biological tissues using mass spectrometry

      Ohno, Timothy R.; Emmons, Matthew; Greife, Uwe; Hofstra, Albert; Lusk, Mark T.; Ranville, James F. (Colorado School of Mines. Arthur Lakes Library, 2019)
      A micro-volume thermo-electronically-cooled trapping device (TEC trap) was designed and built to trap and facilitate subsequent analysis of sub-nanomole quantities of gas. The TEC trap was used with an elemental analyzer - isotope ratio mass spectrometer (EA-IRMS) to provide fully automated carbon-13 analyses of microfossil samples containing less than 100 times material than that required for conventional EA-IRMS analyses. A fully-automated small-sample preparation system incorporating the TEC trap was designed and built to further reduce minimum sample-size requirements for stable isotope analysis without compromising data quality or ease-of-use for large sample batches. This new sample preparation system provided carbon-13 analyses of microscopic animals, more than 100 times below the normal minimum sample-size limits of conventional EA-IRMS equipment while maintaining a precision of better than 0.2 permil. A cryogenically-cooled micro-volume trapping device was designed and built to trap a wider range of gases than the TEC trap and to permit fractional release of the trapped gases. The anticipated outcome of the use of this trap is to allow the analysis of the gaseous contents of a single 10 um fluid inclusion. These devices and techniques have advanced stable isotope research into the microscopic realm and are providing researchers with the ability to investigate individual microscopic organisms and their preserved remains previously beyond the reach of conventional IRMS techniques and equipment.
    • Slope channels on an active margin: a 3D study of the variability, occurrence, and proportions of slope channel geomorphology in the Taranaki Basin, New Zealand

      Plink-Björklund, Piret; Pehlivan, Victoria; Jobe, Zane R.; Wood, Lesli J. (Colorado School of Mines. Arthur Lakes Library, 2019)
      The study of active margin channels has become more accessible with the increased amount of high resolution three dimensional seismic surveys. Channels play an important role in sediment transport from continental region to the sea floor basin. Slope channels are also important reservoirs for petroleum. The Kokako three dimensional seismic survey located west of the Taranaki Peninsula of North Island New Zealand in the Taranaki Basin was used to study Tertiary slope channel development. Eight horizons were mapped and surfaces generated to study the timing of development and morphology of channel systems. Dimensions of channel depth, channel width, and channel length were recorded, and channel sinuosity and width to depth ratios were calculated. Channels found in this region occurred as both individual channels, channel complexes, and as confined channels within a canyon. Stacking patterns observed included components of vertical aggradation as well as lateral migration. Channels on the surfaces had both straight and sinuous morphology. Channel depth ranged from 11 to 393 m, channel width ranged from 65 to 4,462 m, and centerline length of the measured channel ranged from 537 to 47,426 m. Channel sinuosity in the study area ranged from 1.00 to 1.39, and the width to depth ratio ranged from 3 to 63. A number of channel dimensions observed in this study increase in width and decrease in thickness as they move upwards in the stratigraphic column. This trend of increasing width and decreasing thickness during channel evolution indicates that the channels developed such that the base morphology changed to no longer provide accommodation for channel fill, causing a loss of sediment confinement and increase in width. The observed trend in channel aspect ratio and lack of levees indicate that flows responsible for these channels were sheet-like turbidity currents.
    • Optimization of hydraulic fracture treatments and landing zone intervals within the Eagle Ford

      Miskimins, Jennifer L.; Al Mulhim, Abdulrahim Khalid; Ermila, Mansur A.; Miller, Mark G.; Tutuncu, Azra (Colorado School of Mines. Arthur Lakes Library, 2019)
      The current evolution in unconventional reservoirs occurred due to technological breakthroughs in the oil and gas industry. Currently, technologies allow extraction of economical volumes of hydrocarbon from unconventional resources via the assistance of special technologies or massive stimulation treatments. Since new formation development depends on their economics, it is critical to optimize each step within the well stimulation treatments in order to make the development economical. The Eagle Ford shale play has been under development since 2008. Initially, trial and error methods were used to hydraulically fracture wells in the play. These methods provided both good and poor results. This research focuses on optimizing future well landing zones and their corresponding well stimulation treatments within the studied area. Data provided by the Reservoir Characterization Project (RCP) were utilized to develop full geological and geomechanical models using the Grid Oriented Hydraulic Fracture Extension Replicator (GOHFER®) software package. The developed model was calibrated with available field data to ensure that the generated model matches the studied area’s geological and geomechanical characteristics. Base fracture models for two wells, Well B and Well C, were created using the developed geological and geomechanical model. Pressure and production matches for both wells were performed to guarantee that the generated base models reflect what was actually done in the field. The base fracture models were utilized to perform landing zone sensitivity, as well as well stimulation treatment sensitivity analyses. The Austin Chalk and the Eagle Ford formations were examined. Fracturing fluids and their volume, proppant selection, and cluster spacing were sensitized to determine the optimum hydraulic fracture treatment. GOHFER® was utilized to run production analysis for each sensitivity in order to compare the sensitized parameters. Five landing zones within the Austin Chalk and the Eagle Ford were analyzed. The analyses showed that the Eagle Ford had higher oil production potential than the Austin Chalk. 104% and 29% were the increases in the estimated production when the Eagle Ford was targeted instead of the Austin Chalk using slickwater and gel treatments, respectively. Based on the landing zones analyses, the highest oil production, around 326 Mbbl, was obtained from the Pepper Shale formation using a gel treatment. The induced fractures from the slickwater treatment in the Pepper Shale was contained by the Lower Eagle Ford and Buda, while the gel treatment was capable of breaking into the Lower Eagle Ford and accessing additional net pay. Well stimulation treatment sensitivities within the Pepper Shale were also studied. Introducing gel to the slickwater treatment and creating a hybrid treatment improved the oil production; and 70% gel with 30% slickwater yielded the optimum oil production. Larger proppant size performed better than smaller proppant due to conductivity damaging mechanisms. Increasing the fracturing fluid volumes from 175,000 gallons to 300,000 gallons per stage provided negligible increase, around 1%, in the oil production from the Pepper Shale; hence the optimum volume to create the fractures was 175,000 gallons. Thirty foot cluster spacing generated the optimum fracture density and had minimal impact to the production due to stress shadowing. Overall, this research demonstrated that oil production may be increased when the optimized well stimulation treatment is used to hydraulically fracture the Pepper Shale formation.
    • Advancing the computational exploration for thermoelectric materials

      Stevanovic, Vladan; Toberer, Eric; McKinney, Robert W.; Zimmerman, Jeramy D.; Ohno, Timothy R.; Brennecka, Geoffrey (Colorado School of Mines. Arthur Lakes Library, 2019)
      In the computational search for new thermoelectric materials, high-throughput, semi- empirical models have proven to be one of the more fruitful approaches. The best models take into account both electronic and thermal properties since good thermoelectric materials must maintain a difficult balance between the two to achieve high efficiency. In this vein, one route which has proven particularly insightful is to rank materials by their intrinsic thermoelectric quality, which takes into account the density of states, the carrier mobility, and the lattice thermal conductivity, all of which are parameters that either do not change or change in a predictable manner when adjusting temperature or Fermi level. Semi-empirical models of the thermoelectric quality factor (β) have proven successful in suggesting new materials as candidates for good thermoelectric performance. The semi-empirical model of β developed by the Stevanović/Toberer research groups relies on purely isotropic models of the density of states effective mass, carrier mobility, and the lattice thermal conductivity. One aspect which is missing from this approach, however, is any account of the electronic component of thermal conductivity (κe). In thermoelectric materials with very low lattice thermal conductivity, the electronic component can often contribute an equivalent amount to the total thermal conduction. To investigate the potential importance of the electronic component of κ, I developed a novel approach to search for materials with potentially low Lorenz number, which is the coefficient that relates κe to the electrical conductivity, σ. Although the Lorenz number is typically calculated assuming a single parabolic band, I showed that by theoretically driving a material’s energy dispersion away from parabolic, specifically by applying the equivalent of a low pass filter to the energy transport, the Lorenz number can be drastically reduced, leading to a significant enhancement in zT over the single parabolic band approximation. Among the mechanisms by which such an affect could occur are the existence of offset multiple bands in conjunction with intervalley phonon scattering. Based on this plausibility argument, I developed a high-throughput search metric, the density of states shape factor, to provide insight in the search for materials with potentially low Lorenz number. By using this metric as a secondary screening tool in conjunction with the existing semi-empirical β, the vast majority of known thermoelectric materials were found to fall within the search parameters. By extension, new materials within the same bounds were identified for further investigation. In addition to enhancing the search for new isotropic thermoelectric materials, the bulk of my research has been devoted to the development of models to screen materials based on anisotropic transport properties. A computational investigation of anisotropic transport within thermoelectric materials had yet to occur. This absence would have neglected materials which have average isotropic performance, but potentially promising properties along one direction. Well-known thermoelectric materials, such as SnSe, Bi2Te3, and Mg3Sb2, are layered materials in which the intrinsic transport would be inherently anisotropic in single crystals, warranting an investigation into anisotropic transport for single-crystal thermoelectric applications. Before thorough investigations into anisotropic transport began, I conducted a survey of materials which we would expect to demonstrate anisotropic single crystal transport. Layered systems, due to their inherent quasi-2D structures, were obvious candidates. Layered thermoelectrics such as SnSe and Bi2Te3 belong to a class of materials which are bound by loose van der Waals (vdW) bonds. A large set of these vdW layered materials had previously been identified through the use of a slab-cutting algorithm. In addition to the vdW layered materials, there exists another set of layered compounds which is of interest to the thermoelectrics community. Compounds such as Mg3Sb2 and the A1B2C2 Zintl thermoelectrics belong to class of layered materials which are more tightly bound than vdW layered materials, due to the existence of an ionically-bound “spacer” elements between the layers. Using a modified version of the slab-cutting algorithm, that I redesigned specifically for the purpose of finding systems belonging to this class, I identified over 1500 of the so-called “ionic” layered compounds, which are often clays. These compounds exhibit inherent structural anisotropy, yet they are a distinct class from the vdW layered compounds because the bonding between layers can range from very weak, near vdW bonding, to very tight, nearly covalent bonding. Additionally, I was able to show that these materials can be structurally linked to the vdW layered materials from which they can be derived by the addition of a spacer element. By conducting an in-depth analysis of the similarities and differences between these two classes of layered systems and assessing the elastic anisotropy, I revealed a rich diversity of anisotropic behavior within this set, which laid the groundwork for further studies of anisotropic transport on this class of materials. Using both the vdW and ionic layered compounds as a material test set, I began the work of building new semi-empirical models for anisotropic transport. The first step was to create a model of anisotropic thermal conductivity. In addition to thermoelectric applications, a model of anisotropic κL is important for any application in which single crystal thermal conductivity is of interest. I created a new anisotropic model for thermal conductivity which achieved an accuracy within a factor of 2 across 5 orders of magnitude. Applying this model to vdW and ionic layered compounds revealed that anisotropy within κL wanes as the minimum value approaches the amorphous limit. Additionally, by examining the high end and low end of thermal conductivity, new potential materials were identified for thermoelectric or power electronic applications based on their predicted κL(θ,φ). The last part of my investigation into anisotropic transport was to build a new model for the prediction of anisotropic carrier mobility. Building upon intuition from the existing semi-empirical model of mobility, I created the new model for directional mobility by using isotropic and anisotropic elastic parameters along with the conductivity effective mass tensor. By fitting the new model to a set of experimental values gathered for over 60 compounds, the new model achieved an accuracy within a factor of 3 across 4 orders of magnitude, which was a significant improvement over the previous isotropic model. Combining the anisotropic mobility and anisotropic lattice thermal conductivity, I was able to create three different metrics by which to rank materials and screen the vdW and ionic layered compounds to search specifically for materials with ideal anisotropic transport.
    • Interpreting macroscale conductivity behavior of ceria-based oxides via nanoscale quantification of grain boundaries

      Gorman, Brian P.; Diercks, David R.; Burton, George L.; Zimmerman, Jeramy D.; Brennecka, Geoffrey; Kee, R. J. (Colorado School of Mines. Arthur Lakes Library, 2019)
      Non-stoichiometric oxides, which exhibit advantageous electronic and ionic conductivity, are key components in a number of technologically relevant areas including gas separators, solid oxide fuel cells and electrolysis cells. Grain boundaries dramatically limit the charge transport and therefore the overall efficiency of these devices. The limited conductivity is typically attributed to composition and chemistry changes only a few nanometers from these interfaces, due to the different defect formation energies at grain boundaries compared to the bulk. In the following, macroscopic electrical properties of two non-stoichiometric ceria-based oxide systems are related to individual grain boundary compositional and chemical variations, primarily through correlative atom probe tomography (APT) and transmission electron microscopy (TEM). First, a robust technique using the scanning transmission electron microscope (STEM) is developed to automatically analyze grain orientation, a presumed important factor in grain boundary segregation. Second, segregation of oxygen vacancies and cation species were quantified at multiple high angle grain boundaries and at phase boundaries in a dense dual-phase ceramic membrane consisting of BaCe0.8Y0.2O3-δ - Ce0.8Y0.2O2-δ. No trend between misorientation and segregation could be determined. Finally, direct measurements of individual grain boundary composition, electronic structure, and electric potential were systematically investigated and compared between two doping levels in ceria solid solutions: Ce0.99Y0.01O2-δ and Ce0.9Y0.1O2-δ. It was found that the potential was positive for the 1% doped sample, while a negative potential was measured and corroborated by three techniques in the 10% doped sample. While most of the measurements of ceria solid solutions in literature assume a positive grain boundary potential, these results suggest that this is not necessarily always the case.
    • Performance and cost-effectiveness of commercially available adsorptive technologies for treatment of per- and poly-fluoroalkyl substances (PFAS) impacted groundwater

      Bellona, Christopher; Marshall, Robert Eric; Cath, Tzahi Y.; Strathmann, Timothy J. (Colorado School of Mines. Arthur Lakes Library, 2019)
      Per- and polyfluoroalkyl substances (PFAS) are emerging environmental contaminants that have received significant recent attention due to their relatively common occurrence in the environment, recalcitrance in treatment systems and potential adverse human health impacts. While there are currently no maximum contaminant levels established by the United States Environmental Protection Agency for PFAS chemicals, social awareness has motivated the Department of Defense, manufacturing companies, and water providers to address the presence of these contaminants in drinking water supplies by using advanced treatment techniques. Currently, the complexity and costs associated with treating PFAS are barriers for widespread implementation of a remedial practice. Consequently, the primary goals of this study were to evaluate the performance of a novel adsorptive media compared to existing PFAS treatment technologies and develop a decision support tool to aid in PFAS treatment selection based on performance and cost.
    • Interfacial properties of CH₄/C₂H₆ gas hydrate particles with chemical additives

      Koh, Carolyn A. (Carolyn Ann); Hu, Sijia; Zerpa, Luis E.; Carreon, Moises A.; Wu, Ning; Bodnar, Scot; Lederhos, Joe (Colorado School of Mines. Arthur Lakes Library, 2019)
      Gas hydrates are ice-like solid compounds comprised of interconnected cages of water molecules, which can encapsulate guest (gas) molecules. Gas hydrates can form in subsea oil and gas flowlines operating at high pressures and low temperatures, which are the conditions that gas hydrates are thermodynamically stable. The formation, agglomeration, and deposition of gas hydrates can lead to plugging of the oil and gas flowlines, resulting in disruption to production, economic losses, and environmental impacts. The conventional hydrate prevention strategy involves using thermodynamic hydrate inhibitors (THIs), however, high concentrations (e.g., 30-40 vol.%) are often required, which makes this strategy costly and environmentally unfriendly. An alternative strategy is to inject anti-agglomerant (AA) surfactant chemicals into the flowline to reduce hydrate agglomeration by decreasing the hydrate interparticle cohesive forces, while still allowing hydrates to form. In contrast to THIs, the typical effective dosage of AAs is much lower (e.g., 0.5 - 2 vol.%). In this thesis, a novel method has been developed to evaluate and study AA performance to mitigate hydrate interparticle interactions, and subsequent hydrate agglomeration, using a high-pressure micromechanical force apparatus (HP-MMF). These HP-MMF measurements are performed using low sample volumes in the presence of a liquid hydrocarbon phase at high pressure and low- temperature conditions. This new method addresses an outstanding requirement in the hydrate field to provide a quantitative and reliable AA evaluation technique. Model AAs with known molecular structures and commercial AAs (i.e., quaternary ammonium salts) have been studied using the newly developed approach. The results demonstrate that the HP-MMF is capable of accurately capturing the effect of AAs on reducing gas hydrate interparticle interactions (cohesive forces). These HP-MMF measurements are shown to be consistent with the macroscopic apparatuses, such as industrial rocking cells and a pilot-scale flowloop. In addition, the HP-MMF approach for AA evaluation has been extended from synthetic AAs to natural AAs in crude oils. The HP-MMF method has been used to investigate the effect of different parameters on interparticle interactions without AAs as baseline tests, such as the type of continuous phase, annealing (hydrate conversion) time, contact (shut-in) time and salinity. It was observed that the gas hydrate interparticle cohesive force was inversely proportional to the annealing time, whereas the force increased with contact time. For longer contact times (> 12 hr), the force may be too large to be measured since the two hydrate particles would adhere permanently to form one large particle. Furthermore, hydrate interparticle cohesive force in the gas phase is higher than that in the liquid hydrocarbon phase. The addition of AAs can reduce the hydrate cohesive force from the baseline to a non-measurable force with increasing concentration of a high-performance AA (e.g., 0.25 vol.% to 2 vol.% of AA1). An under-dosed scenario was also identified where the hydrate cohesive force significantly increased after long contact periods, eventually resulting in a system failure (irreversible interparticle interactions). On the other hand, the hydrate cohesive force remains non-measurable regardless of the contact time when the AA dosage reaches the MED (minimum effective dosage). With knowledge of the detailed structures of AAs, experiments were conducted to investigate the structure-property effects of AA molecules. It was hypothesized that the AA with a longer alkyl chain length can provide a thicker barrier on the gas hydrate surface (to reduce hydrate cohesive force), and the AAs may have a higher packing density at the hydrate surface when the AA alkyl tail length is comparable to that of the liquid hydrocarbon chain. Increasing the salinity can also reduce or eliminate the effect of particle contact time. The HP-MMF method can capture the effect of the AA tail length and the effect of saturated vs. unsaturated tails on hydrate interparticle cohesive forces. The structure-performance relation approach reported in this work could be used to help improve understanding of the molecular mechanisms and properties of AAs and informed selection and development of new AAs. A possible mechanism is presented in this thesis to describe the effect of contact time on hydrate interparticle cohesive force based on the capillary liquid bridge (CLB) model. A dynamic force model adapted from the CLB equation has been proposed to predict the hydrate interparticle cohesive force as a function of contact time, showing close agreement with the experimental data. To better understanding the hydrate wettability, a key parameter in the CLB model, high-pressure contact angle measurements have been also performed. The data obtained in the liquid hydrocarbon (and the dynamic force model) has been applied to the transport and kinetic models incorporated in multiphase flow software (i.e., CSMHyK coupled to OLGA), and the hydrate risk prediction tool using machine learning. Another outstanding question for the hydrate field is the particle-surface interaction. A gas dominated system may present a higher risk of a hydrate deposit forming on the pipe wall. Therefore, coated pipe surfaces have been applied in this work to reduce the gas hydrate particle-surface interactions in a gas dominated system. It was demonstrated that a superhydrophobic pipe surface can reduce hydrate particle-surface adhesion during a 24 hr test period. Conversely, deposition occurs on an uncoated carbon steel pipe surface in 4 hr.
    • Two dimensional random access multiphoton spatial frequency modulated imaging with at-focus pulse characterization and compression

      Squier, Jeff A.; Allende Motz, Alyssa; Reimanis, Ivar E. (Ivar Edmund); Carr, Lincoln D.; Adams, Daniel (Colorado School of Mines. Arthur Lakes Library, 2019)
      First, we present a newly developed two photon excitation time resolved photoluminescence (2PE-TRPL) microscopy system of high spatial and temporal resolution, capable of detailed 3D analysis for PL emission intensity and minority carrier lifetime. Initial data shows PL intensity, lifetime, and diffusion length imaging in polycrystalline CdTe and indicates variations of diffusion coefficient and bulk lifetime in a polycrystalline CdTe sample. Second, a nonlinear imaging platform is demonstrated with the following novel advances. First, it includes the first implementation of a spectral phase and amplitude reconstruction and compensation (SPARC) platform within a nonlinear microscopy platform. Notably, SPARC is a both second-order compensation system AND provides a means of characterizing the pulse temporal amplitude and spectral phase directly at the microscope focus, allowing for a true characterization of light exposure conditions experienced by the specimen at the image plane. For the first time, to our knowledge, we provide characterization of each of the multi-foci within a random access platform, measured directly at focus. We also provide a Bootstraps error analysis of the pulse retrieval algorithm. We've also demonstrated the utility of the SPARC platform both as a pulse compressor and characterization means. Second, we have shown that spatial frequency modulation imaging (SPIFI), a technique which enables single element detection for extended excitation sources within scattering media, can be extended into two-dimensions with essentially the same mask design used for the original light sheet systems. Here we make use of a spatial light modulator (SLM) in conjunction with SPIFI to create multiple foci at arbitrary points within the image plane, each tagged with a distinct temporal modulation frequency. This enables multifocal imaging within a random-access environment, with 1D detection where the spatial positions are demultiplexed with simple frequency domain processing and provides a pathway towards the enhanced resolution capability of SPIFI in two dimensions for the first time.
    • Neutron induced fission fragment angular distributions and momentum transfer measured with the NIFFTE fission time projection chamber

      Greife, Uwe; Hensle, David; Sarazin, Frederic; Leach, Kyle; Sellinger, Alan; Snyder, Lucas (Colorado School of Mines. Arthur Lakes Library, 2019)
      Nuclear fission is the process by which a large nucleus splits into two heavy fragments and is often induced by an incident neutron. Understanding the probability by which an incident neutron will cause fission, i.e. the neutron induced fission cross section, is an important input into fission applications such as nuclear energy and stockpile stewardship. The Neutron Induced Fission Fragment Tracking Experiment (NIFFTE) Collaboration built a fission time projection chamber (fissionTPC) to study the fission process in a novel way in the hopes of achieving unprecedented precision on fission cross section measurements. With the fissionTPC’s ability to do three-dimensional tracking of fission fragments and other ionizing radiation, systematics of previous cross section measurements using fission chambers can be further explored. Because the fissionTPC records a wealth of data for every fission event, other physics can be measured concurrently with the cross section. In particular, fission fragment angular distributions and the linear momentum transferred from the incident neutron to the target nucleus as a function of incident neutron energy from 130 keV to 250 MeV will be the focus of this work. Angular anisotropy values for 235U and 238U and neutron linear momentum transfer for 235U, 238U, and 239Pu will be presented.
    • Self-reflective experiential learning for persistent autonomy

      Zhang, Xiaoli; Zhou, Xu; Tang, Gongguo; Petrella, Anthony J.; Vincent, Tyrone (Colorado School of Mines. Arthur Lakes Library, 2019)
      Robots have been expected to achieve persistent autonomy for a long time, in which robots are required to safely operate in unknown environments for extended lengths of time while without human interventions. Reinforcement learning holds the promise for persistent autonomy because it can adapt to dynamic and unstructured environments by automatically learning optimal policies from the interactions between robots and environments. However, failures can be unavoidable in the learning process as reinforcement learning can learn the outcome of an action only by executing the action itself. These failures can cause damages to robots or environments in practical applications and hence hinder persistent autonomy. In general, human interventions are usually needed to avoid or resolve the learning failures, but they can be unavailable in practical applications such as space exploration, search and rescue, and underground or underwater construction. Based on a multi-level architecture for persistent autonomy, this dissertation proposes new self-reflective, experiential strategies and methods, aiming at achieving safe adaptions to different environments with minimum human interventions. At the strategic level, while understanding learning is not always necessary and beneficial, this dissertation adds a high-level sophistication of whether and when to learn to reduce failures during learning. At the tactical level, this dissertation proposes a new self-recoverable reinforcement learning algorithm that consists of a multi-state recovery strategy and a failure-prevention strategy. The multi-state recovery strategy improves the learning’s own capability of handling already occurred failures and the failure-prevention strategy learns from failures that are usually ignored and abandoned before to generate a more effective strategy of preventing future similar failures. At the operational level, this dissertation proposes a new multi-objective-optimization-based auto-tuning method to adjust control parameters for robustly achieving the upper-level learning behaviors.