35967 research outputs found
Sort by
INVESTIGATION OF EROSION/CORROSION AND THERMAL-HYDRAULIC BEHAVIOR OF PROTYPIC SODIUM FAST REACTOR COMPONTENTS
No full textThe global search pivot away from fossil fuels has driven a resurgence of interest in the sodium fast reactor, a technology that was used in the original implementation of nuclear energy for power production but has since been overtaken by other reactor designs. Now, sodium fast reactors are one of the six technologies identified as the next generation of nuclear energy. While much research was conducted in the past century, further refinement of our knowledge and experience using liquid sodium as a primary coolant and working fluid within a nuclear reactor will serve to push the industry forward and further the promise of clean, abundant energy.
This thesis describes three areas of research relevant to liquid sodium thermal-hydraulics and material interactions. The erosion/corrosion of stainless steel due to liquid sodium, particularly under high velocity flow conditions, is an area of the existing literature without remaining questions to be answered. Previously conducted experiments indicated that velocity-induced erosion/corrosion has an upper limit, past which increased flow velocity may not have increased effects. To investigate this, a 5000-hour test of erosion/corrosion of six orifices (as would be used for core flow distribution) in prototypical reactor flow conditions was conducted. The results of this test informed the continuation of testing, increasing the oxygen concentration of the liquid sodium flow.
In parallel with corrosion testing of orifices, the hydraulic behavior of the same orifice design was directly compared in sodium and water flows. Often cited as a stand-in for sodium hydraulic behavior, water offers a much more accessible means of testing flow characteristics in design work, since its density and viscosity are similar to liquid sodium and because of the ease of water testing compared to liquid sodium. The loop utilized for corrosion testing was used to perform pressure drop testing of orifices similar to the corrosion test specimens, which were then placed into a
hydraulically identical water test section. This allowed the loss coefficient to be assessed between the two fluids, to confirm that water can effectively be used as a surrogate fluid for predicting sodium flow characteristics where the dimensionless flow characteristic numbers (i.e. Reynolds number) are matched .
Finally, this thesis details thermal-hydraulic performance characterization of a printed circuit heat exchange. These devices are a form of compact heat exchanger offering high surface area per volume and are widely considered for nuclear power applications. There is limited existing research on liquid sodium flow within these devices, and the most comparable study noted uncertainty regarding the heat transfer effects of fluid header regions within the PCHE. To investigate this, a custom designed printed circuit heat exchanger was instrumented with a distributed temperature sensor at multiple points within the device, measuring wall temperature. Operated within a liquid sodium loop and using compressed nitrogen as the other working fluid, this test shed insight into the performance of liquid sodium to nitrogen heat transfer through use of a compact printed circuit heat exchanger. The air-Brayton cycle is considered by several designs for next generation nuclear power, including microreactors. As such, there could be direction application of a sodium-to-nitrogen heat transfer, particularly through a compact heat exchanger.Financial support for several areas of this work was provided by TerraPower LLC, under part of a Department of Energy Advanced Reactor Demonstration Program issued under work order TP-PO-001860, in addition to the University of Wisconsin Thermal Hydraulics Laboratory
GENERATION OF NATURAL LANGUAGE EXPLANATIONS FOR AMCC DERIVED COUNTERFACTUALS
No full textThis thesis explores 2 different approaches to generate simple natural language sentences for counterfactual explanations to better help user understand a machine learning model’s decision. The objective is to determine a domain independent natural language generation approach to generate effective and simple sentences to explain counterfactual explanations. Counterfactual explanations generated from the achievable minimally-contrastive counterfactual explanations (AMCC) algorithm is used to conduct experiments for the two approaches, a template-based approach, enhanced with WordNet synonyms and lexical substitution using a BERT model, and a template-guided approach using a fine-tuned T5 model. Evaluations are performed using an automatic metric, BLEURT, to identify the effectiveness of each approach. To test the effectiveness of domain independence of the proposed approaches, experiments were conducted on a dataset from a new domain to see the effectiveness of generated natural language explanations
Climate Conditioned Season Ahead Streamflow and Agricultural Yield Forecasts for the Upper Colorado River Basin
No full textThe Colorado River Basin provides water to over 40 million people and 5.5 million acres of irrigable land, defining the western United States as I know it today. However, this crucial water supply is currently threatened by a multi-decade drought and cascading demands. These climate driven water scarcity risks are intensified by rapid urban growth in the region; enhanced or alternative supply and demand management practices are desperately needed. Season-ahead reservoir inflow forecasts, coupled with infrastructure management, can inform decisions at lead times of months to seasons or longer. Such forecasts may also provide impetus to shift water use toward high priority sectors and activities. Forecasts from NOAA’s Colorado Basin River Forecast Center are currently used to inform the Bureau of Reclamation’s reservoir modeling system. The River Forecast Center utilizes an ensemble streamflow prediction approach, which performs very well once the high streamflow season begin (typically April); however, longer-lead forecasts, prior to snowpack establishment, tend to be less reliable. Alternatively, incorporating large scale climate variables and teleconnections provides opportunities to improve forecast performance, especially at longer lead times. I present a complementary forecast model framework that explicitly captures asymmetric relationships between climate indices and hydrologic variables. I then investigate the application of this framework to agriculture by evaluating its relationship with annual alfalfa yields, the state's leading crop in terms of economic production value.
Crafts, Contests, and Control?: An Insider Analysis of Bias and Its Effects Within the Cosplay Community
No full textWhen a subculture is built upon the rubble of insecurity, the end product is imbued with the bias and anger of those who laid the first brick. Despite the subculture of cosplay and anime conventions originating from a position of otherness, the community has long felt the sting of bias and bigotry lingering underneath its glimmering petals of chiffon and showmanship. From cosplays very origin, long mistaken to have started with men but was actually birthed by Myrtle R. Douglas, it has suffered under the weight of being seen as a woman's work, despite white, cisgender men taking home many of the awards and recognition rightfully belonging the true crafters - often marginalized groups. Modern day cosplay remains much the same, with frequent attacks on body type, race, gender identity, sexuality, disability, and economic status have permeated the comment sections of photo ops and audiences of masquerades for decades despite recent pushes for acceptance and judgment on craftsmanship alone. Through interviews, historical examination, and my decade of experience, this project seeks to investigate the roots, extent, and damage such attitudes can produce within the cosplay community, from the hallways of the Artist Alley all the way to the judge's table of a competition with thousands of dollars on the line. Consequently, the rules, policies, and criteria surrounding conventions and their craftsmanship contests will be put under a lens to examine whether these negate the damages of bias or if more must be done to regulate a subculture reaching closer to the mainstream
ALTERATION OF MANUFACTURING CONDITIONS TO PRODUCE LOW MOISTURE PART-SKIM MOZZARELLA CHEESE FROM HIGHLY CONCENTRATED MILK
No full textLow moisture, part-skim (LMPS) Mozzarella cheese is a pasta filata style cheese of great commercial importance in the United States due to its frequent use on pizzas. The cheese industry would benefit greatly from improvements to the efficiency of its manufacturing process. Membrane filtration can be used to concentrate milk before cheesemaking in order to achieve increases in cheese production volume. Cheesemilk is often concentrated by up to 1.5×, but manufacturing challenges and cheese quality issues often occur when milk is concentrated beyond this level. We investigated some modifications to the typical manufacturing process for LMPS Mozzarella cheese to allow the use of milk concentrated by 2–4× through microfiltration (MF). We produced LMPS Mozzarella cheese from MF-concentrated milks with varying casein (CN) concentrations between 2.5% (typical levels) and 8.7%. We introduced gentle curd handling procedures and altered the renneting conditions, which helped overcome challenges with rapid coagulation and excessive moisture losses that commonly occur during cheesemaking from concentrated milk. Fat recovery in cheese declined significantly as CN content increased, which resulted in differences in the sensory properties of pizzas made from these cheeses. To address challenges with fat recovery, we then investigated two milk pre-treatment techniques (pre-acidification and homogenization) and one alternative cheesemaking style (a pressed block process compared with the traditional pasta filata process). We evaluated the impacts of these milk pre-treatments and cheese styles on the composition, functional properties, sensory qualities, and microstructure of LMPS Mozzarella cheeses made from 9.5% CN milk
EFFECTS OF STRAIN ON SURFACE DYNAMICS AND ADSORBATE-INDUCED SURFACE RECONSTRUCTION OF METAL NANOPARTICLES
No full textWith advances in characterization techniques, a rekindled interest in the dynamic nature of catalytic materials under realistic reaction conditions has emerged. Specifically, recent studies have demonstrated how under the increased temperatures and pressures relevant for industrial catalysis, adsorbates can induce cluster formation and other structural changes.
First, we investigate the impact of surface strain on adatom formation over eight face-centered cubic (FCC) transition metals using first-principles density functional theory (DFT) calculations. A systematic analysis of both the thermodynamics and kinetics of atom ejection and adatom formation reveals that both tensile and compressive strain destabilize undercoordinated surface atoms and facilitate the adatom formation process.
Second, the role of increased carbon monoxide (CO) coverage was also investigated to further align our models with realistic reaction conditions. It was found that increased CO coverage reduces the energy barriers required for adatom formation by weakening local metal–metal bonds and relieving compressive strain at undercoordinated surfaces. In addition, atomically resolved strain analysis reveals that local strain at reactive kink atoms deviates substantially from the applied global strain and is dynamically reshaped by adsorbate binding. Lastly, we developed structural models to explore equivalent phenomena in cuboctahedral and icosahedral nanostructures, which are advantageous due to their large surface area and abundant undercoordinated sites. The thermodynamics for the least-coordinated atom ejection and adatom formation as a function of CO coverage on icosahedra were examined, as icosahedra consistently showed greater stability compared to cuboctahedra. By precisely optimizing both the active site and structural properties of the catalyst,
these surface models offer a promising strategy for advancing our understanding of reaction mechanisms and provide insights into the nature of active sites in heterogeneous catalysis
The school-to-prison pipeline: Recommendations for early intervention programming for juveniles involved in fights
No full textDisruptive and violent behaviors are not a new problem that school administrators are dealing with. Although, the method by which those behaviors are dealt with has evolved over time. The implementation of exclusionary zero-tolerance policies has increased in recent years. This practice has been found to be more detrimental to students and their learning. The adverse effects of these policies may have also contributed to an increase in referrals to juvenile court services. Students who become involved in juvenile court are less likely to maintain employment in the future and more likely to become further involved in the justice system. Professionals in the field have coined the term school-to-prison pipeline to describe this growing problem. As such, the purpose of this research is to address the ways in which schools can intervene and manage fights and violence at school, while reducing the short and long-term negative impacts on students
INVESTIGATING THE ROLE OF THE MEMBRANE LOCALIZATION DOMAIN (MLD) IN BOTULINUM NEUROTOXIN A (BoNT/A) DURATION OF ACTION
No full textBacteriology MS ThesisBotulinum neurotoxins (BoNTs), produced by Clostridium botulinum, are among the most potent biological toxins known to humans and are the causative agents of botulism, a severe neuroparalytic disease 1. Among the BoNT serotypes (A-G), BoNT/A1 is distinguished by its high potency and prolonged duration of action, whereas BoNT/A3 exhibits reduced potency and a significantly shorter duration 2,3. Preliminary observations suggest that BoNT/A1 and BoNT/A3 may differ in the intracellular localization of their Light Chain (LC) proteases: BoNT/A1 LC predominately localizes to the plasma membrane, while BoNT/A3 LC is distributed throughout the cytosol and on intracellular vesicles 4.
A previously identified region of low sequence homology between BoNT/A1 and BoNT/A3, termed the Low Homology Domain (LHD), contains a subregion implicated in membrane association, designated the Membrane Localization Domain (MLD) 5,6. To investigate the role of the MLD in intracellular trafficking and functional persistence, we engineered recombinant chimeric BoNT LCs and holotoxins by exchanging the LHD or MLD between BoNT/A1 and BoNT/A3. Using LC encoding mRNA transfection or holotoxin intoxication in human induced pluripotent stem cell (hiPSC)-derived motor neurons, we evaluated SNAP-25 cleavage as a functional readout of BoNT activity. Duration of action of holotoxins was further assessed in an in vivo mouse model.
Our results demonstrate that the MLD is essential but not sufficient for the differences in potency and duration of action between BoNT/A1 and BoNT/A3
Teacher Perception and Awareness of the Interrelationship of the Learning Environment Design and Agentic Learning in Public Montessori Settings
No full textMontessori EdDThis exploratory qualitative research study aimed to amplify the voices of public-sector
Montessori teachers by examining their perceptions of the interrelationship between learning
environment design and agentic learning. Using an instrumental case study design, the research
employed semi-structured interviews and photo-elicitation to gather teacher perspectives.
Reflexive thematic analysis was used to interpret key phrases and statements, shedding light on
teachers’ lived experiences, observations, and insights regarding how learning environment
design is related to agentic learning in public Montessori elementary classrooms. This study's
findings suggest that participants understood the interrelationship between the design of the
learning environment and agentic learning. Their descriptions and observations demonstrate an
awareness of their roles and responsibilities in designing the learning environment and an
understanding of agentic learning. The findings of this study contribute to an underexplored area
of study by elevating the perceptions of underrepresented public Montessori teachers and
highlighting their awareness of the interrelationship of learning environment design and agentic
learning