University of Illinois Urbana-Champaign
Illinois Digital Environment for Access to Learning and Scholarship RepositoryNot a member yet
123813 research outputs found
Sort by
Obtaining physical insights for diffusion through machine learning for renewable energy storage applications
Get PDFSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-20 without embargo termsThe student, Grace Lu, accepted the attached license on 2025-07-11 at 11:39.The student, Grace Lu, submitted this Dissertation for approval on 2025-07-11 at 12:49.This Dissertation was approved for publication on 2025-07-14 at 11:35.DSpace SAF Submission Ingestion Package generated from Vireo submission #22508 on 2025-10-20 at 20:15:09Hydrogen storage, oxide fuel cells, and Li-ion batteries are three techniques that enable renewable energy storage and transport. To improve their performance, new materials need to be discovered with ideal transport properties. To explore the large material space efficiently, machine learning methods provide physical insights and enable the rapid screening of new materials. While analytic models use hand-selected features that have clear physical ties, they often lack accuracy when making quantitative predictions. Machine learning models are capable of making accurate predictions, but their inner workings are obscured, rendering it unclear which features are important. Additionally, machine learned interatomic potentials can allow near-DFT levels of accuracy on large systems for which DFT calculations would be prohibitively expensive. To develop interpretable machine learning models to predict the activation energies of hydrogen diffusion in metals and random binary alloys, we create a database and fit six machine learning models. Grouped feature importances, formed by combining the features via their correlations, reveal that the two groups containing the packing factor and electronic specific heat are particularly significant for predicting hydrogen diffusion in metals and random binary alloys. This framework allows us to interpret machine learning models and enables rapid screening of new materials with the desired rates of hydrogen diffusion. We then expand this framework by showing its applicability to predicting transport properties in more complex materials and as a feature down-selection method. For predicting oxygen diffusion in perovskites and pyrochlores, we build a database of experimental activation energies and use our grouping framework to reduce the number of material property features. These features are then used to fit seven different machine learning models. An ensemble consensus determines that the most important features for predicting the activation energy are the ionicity of the A-site bond and the partial pressure of oxygen for perovskites. For pyrochlores, the two most important features are the A-site valence electron count and the B-site electronegativity. The most important features are all constructed using the weighted averages of elemental metal properties, despite weighted averages of the constituent binary oxides being included in our feature set. This is surprising because the material properties of the constituent oxides are more similar to the experimentally measured properties of perovskites and pyrochlores than the features of the metals that are chosen. Inclusion of Ag at the electrolyte-anode interface has been shown to reduce dendrite growth, and enable the use of anode-free lithium-ion batteries through an alloying process. Using the pre-trained MACE-MP-0 potential, we explore interfacial structures between FCC Ag and Li, and conclude that the FCC Li phase is more energetically favorable. We demonstrate that there are negligible migration barriers for Li atoms to migrate across the interface into the Ag. However, larger migration barriers for diffusion of vacancies from the interface into the Li slab can impede the mixing process kinetically
Understanding the complexities of intimate partner violence in Nigeria: exploring the impact on the mother-child relationship, the role of perpetrator's parenting, and the influence of cultural values
Get PDFSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-20 without embargo termsThe student, Tanitoluwa Adeniba, accepted the attached license on 2025-07-11 at 11:51.The student, Tanitoluwa Adeniba, submitted this Dissertation for approval on 2025-07-11 at 11:52.This Dissertation was approved for publication on 2025-07-14 at 11:11.DSpace SAF Submission Ingestion Package generated from Vireo submission #22509 on 2025-10-20 at 20:15:10Despite growing evidence that intimate partner violence (IPV) negatively affects parenting, most research in this area has focused on Western, individualistic contexts, with limited attention to how these dynamics unfold in non-Western, collectivist societies. As a result, little is known about how IPV shapes the caregiving relationship in settings where cultural norms, family structures, and parenting expectations may differ significantly. This study addressed this gap by examining the relationship between IPV and the quality of the mother–child relationship (MCR) among Nigerian mothers. Drawing on coercive control and ecological frameworks, this dissertation assessed whether IPV types, such as coercive controlling violence (CCV) and situational couple violence (SCV), as well as IPV frequency and severity, were associated with MCR quality. It also tested whether father involvement and cultural beliefs moderated these associations. The sample included 101 mothers recruited from community-based organizations in Nigeria. Participants completed an interviewer-administered questionnaire assessing IPV experiences, father involvement, cultural beliefs, parenting stress, and MCR quality. Hierarchical regression analyses were conducted, controlling for parenting stress, child age, child gender, and maternal relationship status. Contrary to expectations, IPV type, frequency, and severity were not significantly associated with MCR quality. Neither father involvement nor cultural beliefs moderated this relationship. However, father involvement and IPV beliefs were associated with poorer MCR quality, and parenting stress was consistently associated with lower MCR quality. The absence of significant associations between IPV and MCR quality may be explained by several intersecting factors, including limited variability in IPV experiences, reliance on maternal self-report, and the uncertain ecological validity of Western-developed measures in the Nigerian context. Additionally, the consistently high MCR scores across the sample suggest that mothers in this high-risk, help-seeking population may engage in protective caregiving strategies to preserve emotional closeness with their children. These patterns collectively underscore the need to move beyond IPV characteristics alone and examine how individual, relational, and sociocultural factors interact to shape parenting outcomes in non-Western, resource-constrained settings
If love were to guide us: the story of Fred Hampton High School’s quest for building bridges of belonging for Black students
Get PDFSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-20 without embargo termsThe student, Dillin Randolph, accepted the attached license on 2025-07-14 at 13:40.The student, Dillin Randolph, submitted this Dissertation for approval on 2025-07-14 at 13:58.This Dissertation was approved for publication on 2025-07-17 at 10:55.DSpace SAF Submission Ingestion Package generated from Vireo submission #22547 on 2025-10-20 at 20:15:13This action research study investigates how Black students experience and make sense of belonging at Fred Hampton High School, a suburban Illinois school known for its racial equity initiatives. Drawing on Critical Race Theory and using Safir and Dugan’s (2021) street data framework, the study centers the voices of Black students through focus group interviews to understand the systemic, interpersonal, and cultural barriers they face. Key themes include racial microaggressions, symbolic leadership, inequitable disciplinary practices, and lack of authentic representation. Despite the school’s reputation for diversity and strong racial equity work, the findings demonstrate the factors that explain why Black students report the lowest sense of belonging among all racial groups at Fred Hampton High School. Contributing factors include performative equity practices and administrative inaction. This study highlights students’ critiques of Hampton’s school culture and offers practical, student-informed recommendations for fostering an inclusive racial climate. By using storytelling as both a methodological and theoretical tool, this research aims to disrupt deficit narratives, promote student agency, and reimagine what is possible when school transformation is guided by love
Detection of refrigerant leaks with focus on low-GWP mixtures
Get PDFSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-20 without embargo termsThe student, Yile Xu, accepted the attached license on 2025-07-16 at 16:03.The student, Yile Xu, submitted this Thesis for approval on 2025-07-16 at 16:18.This Thesis was approved for publication on 2025-07-23 at 14:51.DSpace SAF Submission Ingestion Package generated from Vireo submission #22612 on 2025-10-20 at 20:15:22In the context of global climate goals and tightening environmental regulations, the use of low-global-warming-potential (low-GWP) refrigerants, particularly flammable A2L-class and zeotropic blends, is rapidly increasing across refrigeration and air-conditioning systems. This is a comprehensive study of zeotropic refrigerant leakage under different scenarios. The study begins with a detailed literature review on refrigerant leakage effects, detection principles, and physical leakage modeling, establishing the background for subsequent research. Experimental facilities were designed to test thermal conductivity-based sensing, including both steady-state and gas-flow leakage scenarios, as well as a pipeline leakage test platform simulating realistic HVAC system conditions. Methods for calibrating flowmeters and estimating key parameters were developed to ensure measurement accuracy. The experiment on refrigerant leakage in gas flow and in a sealed chamber were conducted, and another experiment of refrigerant leakage from pipeline is underway. The experimental results demonstrate that modern thermal conductivity sensors provide more reliable responses to preset refrigerant concentrations, though actual readings tend to slightly underestimate the expected LFL percentages. In contrast, Pellistor-like sensors (e.g., VQ31MB) showed selective responsiveness, detecting R1234yf but failing to respond reliably to blended refrigerants, suggesting limited applicability in complex leak scenarios. Additionally, by analyzing diffusion dynamics across refrigerants, the work highlights the influence of molecular weight, structure, and density on dispersion rates, offering critical insights for sensor placement and calibration strategies. Leakage modeling and preliminary simulation efforts complement the experimental work, paving the way for future studies to optimize system-level detection strategies. Looking ahead, the research identifies key areas for further exploration, including real-world leakage validation, localization of leak sources, system failure time assessments, and multi-fault scenario analysis. Collectively, this thesis advances the understanding of refrigerant leakage detection, providing both practical insights and a methodological foundation for improving the safety and sustainability of next-generation refrigeration systems...
Navigating college choice: the influence of parental education, advice source, and anticipated belonging
Get PDFSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-20 without embargo termsThe student, Ravon Pittman, accepted the attached license on 2025-07-20 at 20:32.The student, Ravon Pittman, submitted this Thesis for approval on 2025-07-20 at 20:56.This Thesis was approved for publication on 2025-07-22 at 16:40.DSpace SAF Submission Ingestion Package generated from Vireo submission #22669 on 2025-10-20 at 20:15:32This study examined how students' anticipated sense of belonging at prospective postsecondary institutions is associated with their parents’ highest level of education, and whether this relationship is moderated by the source of college advice. Data from a diverse sample of high school students in Illinois (N = 234) were analyzed using linear regression models. Results showed parental education alone did not significantly predict students anticipated belonging. Students who relied on formal advice sources (counselors, teachers, mentors) reported significantly higher anticipated belonging at match schools compared to those who received informal advice (family or peers). However, no significant interaction was found between parental education and advice source. Results highlight the importance of formal advising in strengthening students’ academic self-perception and sense of institutional fit. Implications for reducing college undermatch and expanding equitable access to supportive advising networks are discussed
Computing over in-vitro predictive coding neural cultures
Get PDFSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-20 without embargo termsThe student, Shrusti Jain, accepted the attached license on 2025-07-22 at 20:52.The student, Shrusti Jain, submitted this Thesis for approval on 2025-07-22 at 20:58.This Thesis was approved for publication on 2025-07-23 at 09:43.DSpace SAF Submission Ingestion Package generated from Vireo submission #22701 on 2025-10-20 at 20:15:41While artificial neural networks are gaining in popularity, they still fall behind biological neural substrates in terms of energy efficiency and performance on certain computational tasks. However, the mechanisms by which neurons operate are still not well understood, making it difficult to harness their computational power for arbitrary tasks. Predictive coding is an influential theory of learning and inference within neuroscience, positing that neural systems adapt to best predict sensory input across time, thus representing the sensory distribution within an internal generative model encoded through synaptic connections. However, previous work on predictive coding has been limited to modeling relations between higher-level units of the brain. Here, we present a biologically plausible model of predictive coding generalized to arbitrary topologies of in-vitro cultures. In addition, we present a novel framework to harness neural cultures that implement predictive coding for computational tasks
Phase field model to study the corrosion of structural alloys in the harsh environments
Get PDFSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-20 without embargo termsThe student, Harsha Pandey, accepted the attached license on 2025-07-23 at 12:51.The student, Harsha Pandey, submitted this Thesis for approval on 2025-07-23 at 12:57.This Thesis was approved for publication on 2025-07-23 at 15:32.DSpace SAF Submission Ingestion Package generated from Vireo submission #22708 on 2025-10-20 at 20:15:42Ni- and Fe-based structural alloys used in Molten Salt Reactors (MSR) are susceptible to corrosion when exposed to molten salts. Even though the chemical reaction between structural alloys and the main constituents of fluoride and chloride salts of interest to reactors are not thermodynamically favored, the presence of impurities within the salt can initiate the oxidation and dissolution of Cr-content in structural materials. Additionally, machining irregularities in the alloys with added mechanical and thermal stresses synergistically can lead to environmentally assisted cracking (EAC) phenomena, including stress corrosion cracking (SCC), undermining the durability of structural components. This combination of environmental factors can prove to be detrimental to the health of the structural alloys posing a risk to the safety of nuclear reactors. The multi-physics nature of EAC phenomena and the operational complexities in working with molten salts make it challenging to test integral effects experimentally and motivate the development of a computational model that can support and enhance experiments. In this endeavor, modeling of structural alloy corrosion under external stresses in MSRs is realized with the phase-field methodology (PFM) in MOOSE because of the ability of this approach to successfully capture the multiphysics nature of the problem and eliminate the mesh dependencies and displacement discontinuities present within other Finite Element Methods by regularizing the solid-salt interface. Metal dissolution and pit propagation are modeled by the Kim-Kim-Suzuki (KKS) model which is employed to minimize the free energy of the system and take into account the contribution of electrochemical energies as well as mechanical energies due to the applied loads. This thesis will present the modeling approach developed, its validation strategy, and the next step to study structural alloy corrosion in FLiBe salt
Mitigation of simultaneous switching noise by active rail clamps
Get PDFSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-20 without embargo termsThe student, Alec Wasowicz, accepted the attached license on 2025-07-23 at 18:36.The student, Alec Wasowicz, submitted this Thesis for approval on 2025-07-23 at 18:42.This Thesis was approved for publication on 2025-07-25 at 09:13.DSpace SAF Submission Ingestion Package generated from Vireo submission #22718 on 2025-10-20 at 20:15:44In integrated circuits with large amounts of digital switching, conditions may exist such that a significant amount of simultaneous switching noise (SSN) will be present. Low-speed products such as microcontrollers use less advanced packaging technologies, increasing the amplitude of the voltage excursions created by SSN. Overvoltages can lead to reliability issues that degrade transistor lifetime. This work investigates if the overvoltages due to SSN can be mitigated using rail clamp circuits developed for on-chip protection against electrostatic discharge (ESD)
Fertilizer source and placement affect nutrient mobility through the soil profile and productivity of maize
Get PDFSubmission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-20 without embargo termsThe student, Gabriela Frigo Fernandes, accepted the attached license on 2025-07-23 at 18:56.The student, Gabriela Frigo Fernandes, submitted this Thesis for approval on 2025-07-23 at 19:20.This Thesis was approved for publication on 2025-07-25 at 10:20.DSpace SAF Submission Ingestion Package generated from Vireo submission #22719 on 2025-10-20 at 20:15:44Maximizing maize (Zea mays L.) grain productivity for high-yielding areas requires aligning the nutrient availability in the soil with the crop nutrient demand throughout the season. While the traditional practice of pre-plant broadcast fertilizer application is widely used due its convenience, it may be limiting nutrient availability and grain yield due to nutrient tie-up, loss, and/or positional unavailability. The mineral nutrients phosphorus (P) and potassium (K) are considered immobile in the soil and are typically thought to have limited movement through the soil profile. Concentrating these nutrients, however, may enhance their movement into the soil profile and increase their availability for root uptake. Complementing a traditional fertilizer program by adding the multi-nutrient fertilizer source POLY 4 (containing K, S, Ca, Mg) that contains S, and including biostimulant coatings like sugars or humic acid (HA) may further improve nutrient availability and crop productivity. For these reasons, the objective of the first chapter was to investigate how different timing × placement (pre-plant broadcast or in-season surface dribble, also known as dry-drop) and sources of fertilizer, with and without coatings, affect soil nutrient availability, plant uptake, and productivity of maize. The focus of the second chapter was to determine how fertilizer placement affects movement of nutrients into the soil profile and their subsequent availability throughout the crop season. The movement of P, K, and S into the soil profile, and the subsequent impact on maize productivity was examined by comparing the traditional P and K fertilizer sources with added POLY4, with or without organic coatings, and by pre-plant broadcasting the fertilizer or by concentrating it along the crop row in-season as a dry-drop application. The second chapter examined the use of traditional pre-plant broadcast of P, K, and S fertilizers compared to two methods of in-season concentrated surface applications as a granular dry fertilizer (dry-drop) or as liquid fertilizer (Y-drop). Results for soil characteristics varied by site and year, and P, K, and S exhibited stratification patterns across treatments. However, when those nutrients were concentrated in one spot, by dry-drop or Y-drop placement, they consistently increased near-row soil concentrations of P, K, and S compared to pre-plant broadcast, with an average increase of + 32 mg kg-1 in soil P, +43 mg kg-1 increase in soil K, and +12 mg kg-1 increase in soil S. Concentrating the nutrients moved P and K through the soil profile down to as much as 30 cm depth, challenging the conventional idea that these nutrients are immobile. In 2023, less-than-average seasonal rainfall slowed nutrient solubilization and incorporation, reducing the effectiveness of the in-season fertilization, while 2024 experienced more typical rainfall, resulting in enhanced nutrient availability, crop nutrient uptake, and growth compared to 2023. Even though soil tests indicated nutrient sufficiency at most sites, fertilizer applications resulted in measurable yield increases across all the fertilizer treatments. Despite minimally impacting early vegetative growth, fertilizer applications containing POLY 4 significantly enhanced grain yield, producing the highest yield of 13.44 Mg ha⁻¹. The in-season applications were year- and site-dependent, but the overall results in grain yield were at least equivalent to pre-plant broadcast application. The humic acid and sugar coatings did not increase yields, and the combination of humic acid and sugar coating tended to reduce yield by up to 0.32 Mg ha⁻¹. Tissue analysis indicated that grain nutrient concentrations were mostly unaffected by the fertilizer treatments, while stover exhibited increased K and S accumulation when provided with POLY 4. This study supports the hypothesis that in-season surface banding of fertilizer near the crop row increases nutrient availability in the proximity of the crop root system. Fertilizer application increased grain yield in most of the experiments, regardless of the area being considered nutrient sufficient based on soil test recommendations. Overall, this research highlights the importance of fertilizer placement and fertilizer source for modern maize production systems
Modeling of transport mechanisms and quality changes during the microwave frying of foods by solving hybrid mixture theory-based unsaturated transport equations coupled with maxwell's equations of electromagnetism
No full textSubmission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2027-08-01The student, Yash Shah, accepted the attached license on 2025-05-17 at 08:45.The student, Yash Shah, submitted this Dissertation for approval on 2025-05-17 at 09:19.This Dissertation was approved for publication on 2025-05-23 at 11:20.DSpace SAF Submission Ingestion Package generated from Vireo submission #22301 on 2025-10-21 at 10:05:17Microwave frying (MF) has emerged as a promising alternative to conventional frying (CF) to produce healthier fried foods with lower oil content. However, the limited understanding of the transport mechanisms involved in MF can hinder process optimization efforts. Porous media modeling can help address this gap. In this research, the MF of foods was modeled by solving hybrid mixture theory-based two-scale unsaturated transport equations and Maxwell's equations of electromagnetism. The food matrix was modeled as a deformable and viscoelastic material. Modeling the MF of foods is challenging due to the multiple phases (gas, water, oil, and solids) and physics (heat transfer, mass transfer, deformation, and electromagnetics) involved in the process. A stepwise approach was employed wherein the model complexity was increased in each step. First, a previous CF model developed in our research group was modified and solved to account for viscoelastic deformations. Then, the microwave drying of foods was modeled as it is simpler than MF due to the absence of the oil phase. Finally, MF was modeled. Frying and microwave drying experiments were conducted to collect the model validation data. While earlier frying models in the literature ignored the volume changes of foods during frying, the CF model developed in this research accounted for the deformation of the food matrix by utilizing the pore pressure (p_pore) as the driving force governing deformation. The negative gauge p_pore near the sample surface during frying was expected to have caused the contraction of the surface layers, and the potato sample shrank by 18.5% for a frying time of 300 s. The p_pore is also expected to impact the oil penetration in foods during frying. The oil content of the sample increased significantly in the first minute of frying when the p_pore in the food was low. The oil content profile plateaued in the intermediate frying stages. This was expected due to an increase in the magnitude of p_pore in the sample. The p_pore attained a peak value of 19.2 kPa (gauge) at the sample center. The subsequent decrease in p_pore was expected to have enabled the oil uptake by the sample in the later frying stages. During microwave drying, the electric field was centrally concentrated in the cylindrical potato sample. However, the heat concentration behavior was size-dependent. The high magnitudes of pressure in the sample core (peak gauge p_pore at the center: 103.8 kPa) during microwave drying caused outward moisture movement and an expansion of the core. The magnitude of microwave power dissipation decreased in the drier parts of the sample, which may help avoid internal burning and aid the 'moisture-leveling' effect of microwave drying. Sensitivity analysis revealed a significant impact of changes in microwave frequency on the drying of foods. Experiments showed that MF at 2.45 GHz frequency led to the highest heating rates and pressure magnitudes (peak values at the sample center: 107.3°C and 24.9 kPa), followed by MF at 5.8 GHz frequency (peak values: 104.1°C and 20.8 kPa) and CF (peak values: 100.5°C and 13.8 kPa). Below a moisture content value of 3 g/g solids, the oil content of French fries increased rapidly with a decrease in moisture content for CF and relatively slowly for MF. The stress relaxation data of French fries indicated that MF at 5.8 GHz can produce crunchier fries, potentially due to intense crust heating at this frequency. Simulations with the MF model showed that MF produced samples with lower oil content than CF at a given endpoint moisture content (3-33% reduction in oil content). MF also reduced frying times by 33-76%. The spatiotemporal distribution profiles for variables like electric field, microwave power dissipation, temperature, pressure, and moisture content were analyzed. The power dissipation followed the drying front in the French fries during MF. This can cause the thickening of the crust and make the French fries crunchier. MF at 2.45 GHz led to samples with lower oil content than MF at 5.8 GHz, likely due to the deeper penetration of microwaves and higher magnitudes of p_pore in the sample during the former. Processing implications were discussed, and suggestions were made for future improvements to the design of the microwave fryer prototype used in this research. The results from this work improved the understanding of the transport mechanisms involved in MF