Open Research Oklahoma (Oklahoma State Univ.)
Not a member yet
42035 research outputs found
Sort by
Lithium-ion battery fires in commercial aircraft
No full textLithium-ion battery fires on aircraft have increased 178% from 2016 to 2024 (FAA, 2025). These batteries presented a critical safety challenge due to the potential for thermal runaway that could result in fire, explosion, and toxic gas release. Current Federal Aviation Administration (FAA) Regulations permit passengers to carry one 300 Wh mobility aid battery and two 160 Wh spares, totaling 620 Wh per person. However, limited data exist regarding the effectiveness of onboard extinguishing systems against such high-energy failures. This study aims to evaluate the performance of standard aircraft fire suppression methods when applied to lithium-ion battery thermal runaway events. Preliminary testing involved inducing mechanical and thermal failures in common personal-use and mobility aid batteries. This characterized the fire behavior and explosion risk of the batteries. Results showed that large 300 Wh batteries produced 24-inch flame jets and high-velocity shrapnel, posing severe hazards to passengers, crew, and cabin materials. Building on these findings, a full-scale experiment including five tests to assess how a larger mobility aid watt-hour battery failure affected aircraft components - including luggage, seats, overhead bins, Passenger Service Units, and oxygen generators. Standard onboard extinguishing procedures using Halon 1211 and water were applied. The outcome of this test found that current onboard extinguishment methods were insufficient in suppressing fires from 300 Wh lithium-ion mobility aid batteries. These results will be used to guide future onboard safety protocols across the commercial airline industry. It may also influence future FAA policy regarding allowable lithium-ion battery capacity on commercial flights
Addressing barriers of adverse childhood experiences in childcare
No full textAdverse Childhood Experiences (ACEs) are a significant public health concern, with nearly two-thirds of adults in the United States reporting at least one ACE before the age of 18. These experiences-ranging from abuse and neglect to household dysfunction have been linked to long- term impacts on physical health, mental well-being, and socioeconomic outcomes. This thesis explores the correlation between the availability of high-quality childcare facilities and the prevalence of ACEs at the county level in Oklahoma. The study investigates whether increasing the number of childcare centers that offer routine mental health assessments can contribute to a reduction in reported ACEs. Findings aim to highlight the need for a more proactive approach in early childhood environments, particularly in underserved communities, to ensure early identification, intervention, and connection to support systems. Addressing these barriers through systemic investment in childcare infrastructure and culturally competent mental health services is essential to mitigating the long-term effects of childhood trauma
Accuracy and effectiveness of virtual fencing in production settings
No full textVirtual fence (VF) technology can potentially enhance profitability, animal welfare, and environmental impact by replacing physical fences, reducing labor, mitigating wildlife conflicts, and protecting sensitive areas. Virtual fence technology is successful in research environments; however, little is known about the factors contributing to success in production settings.
As VF depends on GPS locations, understanding the accuracy of the GPS-enabled collars is crucial. We evaluated the accuracy of VF collars over 21-days using the accuracy measurements circular error probable (CEP), 95% radius (R95), distance root mean squared (DRMS), twice the distance root mean squared (2DRMS). These accuracy measurements revealed that closed canopies are less accurate despite reporting similar totals of GPS locations (P < 0.01). Collars also exhibit differences in directional distribution of GPS locations depending on the canopy type (χ <0.01).
The initial introduction to VF plays a large role in the success of the technology implementation. Evaluation of four individual training protocols and surrounding factors identified how each influences the effectiveness of training cattle to VF. Through a backwards stepwise regression analysis, we identified that a training period is necessary to decrease the use of excluded areas (P < 0.01). The final model identified three key variables that influenced the use of exclusion zones: training, exclusion area size, and interactions with the VF during evaluation (AIC = 780.56).
Finally, we explored the use of VF in creep grazing and compared it to traditional electric fencing (EF). VF calves accessed creep grazing areas more frequently than EF calves, resulting in a greater average daily gain (P = 0.03). Virtually fenced calves spent 18% of daily GPS locations in the creep grazing area, while EF calves spent 7%. However, cows in VF pastures did access the pasture, so this management strategy was 86% effective in preventing cows from accessing the calf creep grazing area.
These studies improved our understanding of VF technology's accuracy, limitations, and benefits, which is crucial for successful implementation. Further research is needed to optimize collar accuracy and training protocols. Overall, VF has the potential to positively impact the environment, improve grazing practices, and replace traditional fencing
Effect of individual and multiple sensory interruptions on psychological and behavioral outcomes
No full textIn this dissertation, I take an experimental approach to investigate how visual and auditory sensory interruptions affect psychological and behavioral outcomes. In study one, I use a laboratory experiment to test the effects of interruption type (auditory or visual), interruption delivery order, and interruption delivery channel (virtual or environmental) on psychological (self-efficacy) and behavioral (task score and completion time) outcomes. I found that the order and type of interruption affect psychological and behavioral outcomes when delivered via the environment but not for virtual interruptions. In study two, I further investigated interruptions in an exclusively virtual setting and added multiple interruptions to explore the longitudinal effects of the order and frequency of sensory interruptions, adding exhaustion as a psychological outcome to assess the longitudinal effects. I found that exhaustion increased with exposure to multiple interruptions and found a desensitization effect for the dependent variables. Overall, I found that while environmental interruptions have a negative effect, virtual interruptions have no negative effect on behavioral or psychological outcomes, even with the use of multiple interruptions
Examining the fungal community found in iguania digestive tracts
No full textAnaerobic gut fungi (AGF; phylum Neocallimastigomycota) contribute to the degradation of plant biomass within the digestive tracts of herbivores. Regardless of their ecological importance, their evolution and host range remain poorly understood. Most studies, both culture-dependent and independent, have focused on herbivorous mammals, with limited research on AGF in reptiles. Comparative molecular dating of AGF from tortoise and mammalian feces suggests AGF predate their known mammalian hosts, with reptiles as potential original hosts. Beyond tortoises, herbivorous reptiles in the infraorder Iguania (squamate reptiles like iguanas, agamids, and other lizards) are capable of hindgut fermentation. With a prolonged solids retention time and their mostly plant-based diet, Iguania might be suitable AGF hosts.
To test the hypothesis that Iguania host evolutionary older AGF, we sampled wild and zoo-housed species within this order. We employed amplicon sequencing targeting the D1/D2 region of the LSU region using both general fungal and AGF-specific primers. Our objective is to identify and characterize the AGF communities within this order and place it within the context of their broader fungal communities. Preliminary results show a variety of known and novel AGF present. A more detailed data analysis is currently underway. Additionally, quantitative PCR will be used to estimate AGF abundance relative to total fungal abundance in these samples.
This study aims to provide a comprehensive understanding of AGF communities in herptiles, shedding light on the ecological roles and evolutionary history of AGF. It is the first detailed report on the AGF and broader fungal community in Iguania
Two-level ensemble approach for diabetic retinopathy prediction
No full textDiabetic retinopathy (DR) remains a leading cause of blindness among the working-age population globally, necessitating early detection to mitigate its impact. Traditional diagnostic methods, such as direct ocular examinations, are often resource-intensive and inaccessible in low-resource settings. This thesis proposes an advanced predictive model for DR based on routine laboratory results applying a multi-level ensemble approach with hyperparameter tuning to develop an accessible, cost-effective, and accurate model. By integrating routine lab results into the predictive process, this research aims to improve the early detection of DR, particularly in low-resource environments, thereby enhancing patient outcomes and reducing the global burden of vision impairment due to diabetic retinopathy.
The study employs an ensemble of simple, interpretable machine-learning algorithms: Random Forest, Gradient Boosting, Linear SVC, and Extreme Gradient Boosting. Each algorithm undergoes hyperparameter tuning, evaluating various loss functions, scoring methods, and solvers to optimize performance. Using stacking as an ensemble technique, the model combines the strengths of individual algorithms, enhancing predictive accuracy and robustness while mitigating the limitations of single-model approaches.
The proposed model's simplicity ensures interpretability, a crucial aspect in healthcare, allowing clinicians to understand and trust the model's predictions. Moreover, the reduced computational complexity facilitates quick processing of large datasets, an essential requirement in medical settings. The performance of the proposed model will be compared with deep learning models, such as Fully Connected Networks (FCN), other ensemble models, one-level ensembling, and all the individual models used in this research to validate its effectiveness
Relationship between diverse student interactions and leadership development
No full textRecently, legislators have called into question diversity, equity, and inclusion (DEI) efforts among higher education institutions (HEIs) in the United States, resulting in decreased support or the elimination of DEI efforts. Previous research on the benefits of diversity interactions on college student outcomes has often privileged White students and found a positive relationship. The present quantitative study expands on previous research by utilizing correlation, independent samples t-Tests, analysis of variance, and multiple regression statistics to analyze the relationship between interracial diversity interactions and student leadership development among 168 undergraduate students who self-identified with a minoritized race at HEIs in the United States. Positive diversity interactions were found to be positively associated with leadership development. Although not a significant relationship, negative diversity interactions were inversely associated with leadership development. Additionally, diverse participation across the collegiate experiences examined in this study was found to be a significant predictor of leadership development. The results indicate that creating a positive environment for students to interact across racial groups, including diversifying involvement across collegiate experiences, and mitigating negative interactions is important to the development of minoritized students’ leadership development. As the landscape of higher education continues to shift, the role DEI efforts play in HEIs becomes increasingly important to higher education leaders as they produce new leaders that graduate and contribute to the global world
Exploring radiation damage in halide perovskites through computational methods
No full textUnderstanding the behavior of materials under irradiation is crucial for the design and safety of nuclear reactors, spacecraft, and other radiation environments. One key parameter in assessing radiation damage is the average threshold displacement energy (Ed), which quantifies the minimum transferred kinetic energy required to displace an atom from its lattice site, generating a permanent defect. We calculate Ed using ab initio molecular dynamics (AIMD) for lead halide perovskites MAPbI₃, FAPbI₃, and CsPbI₃. The Ed values we obtain, which are considerably lower than those commonly assumed in the literature for several species, hold significant implications for predicting radiation damage to these materials. We perform Monte Carlo simulations using the software Stopping and Range of Ions in Matter (SRIM) with the AIMD-obtained Ed and compare them to simulations using the default Ed. Our results show an increase in specific vacancies in halide perovskites that is not captured in simulations using the default Ed values. This suggests that the default Ed values may not accurately predict the radiation damage in these materials, potentially leading to underestimation of the damage. These results are discussed in the context of the radiation hardness of the materials and suggest they get damaged when exposed to radiation, but they are resilient due to self-healing processes.
Additionally, to complement this work we calculate the electronic stopping power (Sₑ) curves for a hydrogen projectile traveling through two perovskites, FAPbI₃ and FA(Pb₀.₈₇₅Sr₀.₁₂₅)I₃, using first-principles time-dependent density functional theory (TD-DFT) and SRIM. Both methods yield similar results, indicating that 12.5% Sr doping in FAPbI₃ does not significantly impact the unchanneled Sₑ. TD-DFT simulations provide detailed insight into the initial interactions between the projectile and the perovskite material, enabling a comparison of the roles of anions and cations in radiation processes for both pristine and Sr-doped FAPbI₃.
Finally, since traditional methods for determining Ed involve computationally expensive simulations or complex experiments. This project proposes a novel approach leveraging machine learning techniques, specifically the Sure Independence Screening and Sparsifying Operator (SISSO) method, to develop analytic expressions for predicting the average Ed in a wide variety of materials
A scalable process for extraction and macroporous resin enrichment of podophyllotoxin from eastern red cedar (Juniperus virginiana L.)
No full textPodophyllotoxin (ptox) is a plant-derived pharmaceutical used for medicinal treatments from inflammatory diseases to cancer. Originally sourced from the rhizomes of the Indian mayapple (Podophyllum emodi Wall.) and leaves of the American mayapple (Podophyllum peltatum L.), ptox is also found in eastern red cedar (ERC; Juniperus virginiana L.) foliage. This work demonstrated a scalable process to extract ptox from ERC field-dried foliage that was rehydrated and flaked. Optimum extraction conditions were 20 % ethanol, 60° C, 20:1 solvent: feedstock ratio for 1 h, followed by filtration to yield 5.27 mg ptox/g dry foliage. Extract ptox was enriched using three hydrophobic macroporous resins varying in particle size. The small particle size resin (PCG900M; 50–100 µm) exhibited the highest ptox binding capacity (57 mg ptox/g resin) and solvent reduction (11–12 fold), followed by the mid-sized resin (PCG900C; 100–200 µm, 35 mg ptox/g resin, 8 fold solvent reduction), and was lowest for the large particle size resin (PAD900; 300–1200 µm, 29 mg ptox/g resin, 6–7 fold solvent reduction). Conversely, desorbed ptox purity was greatest for PAD900 (19.7 %), intermediate for PCG900C (14.8 %), and lowest for PCG900M (13.1 %). Ptox recovery from ERC extracts was lower for PAD900 (60.0 %) than for both PCG900C (87.3 %) and PCG900M (86.9 %). This optimized process for extraction and enrichment of ptox from ERC foliage offers an incentive for removing ERC trees invading land in rural and increasingly populated areas of the United States.Horticulture and Landscape ArchitectureBiosystems and Agricultural Engineerin
Assessing single-pilot operations: The critical aviation safety data gap
No full textTechnological advancements throughout aviation history have progressively decreased the number of flight crew necessary to operate a commercial airliner safely. While short- and medium-range airliners sometimes operated with two pilots, larger long-range aircraft required up to five crew members until technology and automation replaced the safety margin those roles provided. With the introduction of the first single-pilot business jet in 1977 and nearly 50 years of operational experience later, it remains challenging to determine how the safety of these aircraft compares with that of dual-pilot aircraft under similar conditions. Even with the advancements in this technology, single-pilot operations are allowed under Parts 91 and 135 but prohibited under Part 121. Proponents of single-pilot operations argue that it would save airlines and consumers money and, with advanced technology and training, provide a level of safety comparable to that of two-pilot operations. Those opposed argue that it would eliminate a key redundancy of the current system: having two qualified pilots on the flight deck at all times. This paper examines the historical evolution of flight crews and their roles, the lack of data on single-pilot-certified aircraft in business aviation, and the available research regarding single-pilot operations. While future technological advancements and their proven effectiveness may support further cockpit automation, the paper argues that there is currently insufficient peer-reviewed evidence and safety data to justify an expansion to Part 121 at this time