Master's Theses (Fall 2009 to Present)

Permanent URI for this collection

https://udspace.udel.edu/handle/19716/5010

New submissions to the University of Delaware Master's Theses collection are added as they are released by the Graduate College. The Graduate College deposits all master's theses from a given semester after the official graduation date.

University of Delaware master’s theses submitted between 1980 - Summer 2009 are available online through Dissertations & Theses @ University of Delaware. Use the library catalog, DELCAT, to search for all print or microform copies of master's theses 1980 - 2009 that are NOT available in Dissertations & Theses @ University of Delaware because Dissertations & Theses @ University of Delaware does NOT contain the complete collection of University of Delaware master's theses.

Master’s theses in the Longwood Graduate Program in Public Horticulture (now known as the Longwood Fellows Program) submitted between 1970 - 2004 are available online.

More information is available at Dissertations & Theses.

Browse

Now showing 1 - 20 of 2432

Assessment of seabed munition mobility induced by propeller-generated turbulence
(University of Delaware, 2025) Swanson, Christopher Q.
In the years prior to the contemporary environmental movement, navies around the world dumped unused munitions overboard for disposal. In 1972, the London Convention banned many pollutants, including munitions. Half a century since, these potentially live munitions can come onshore or be uncovered by erosion, posing serious risks to the public. Among the mechanisms driving munitions mobility, this study investigates the influence of propeller generated turbulence on munitions transport. The experiment used a large pool with a sediment bed and a proud munition model (also referred to as target). An overhead outboard motor was used to simulate vessel propeller wash. Results showed munition density was a parameter governing target stability under propeller-generated flows. Targets constructed to emulate high density conventional munitions demonstrated strong resistance to displacement by not exhibiting movement under tested flow regimes. In contrast, movement was observed only when the targets were configured to be less than half as dense as conventional ordnance. Detailed measurements further established that the magnitude of target displacement was governed by density and sensitive to spatial proximity to the propeller; targets positioned near the point of maximum jet velocity experienced substantially larger mobilization. The patterns affirm that, in deep-water scenarios where propeller-induced turbulence dissipates rapidly and munitions remain dense, the overall risk of large-scale munition migration is low. Nevertheless, there is a heightened vulnerability for less dense munitions or scenarios characterized by high-energy, confined settings—such as navigation channels and high traffic port facilities—where persistence of strong, localized flows can present a mobilization threat. The ability to predict munition movement, combined with general historical disposal data are crucial to predict when/where expected munitions may arise. Local authorities can then exclude the populace until ordnance experts can properly dispose of them in a safe and formal process, akin to the Iron Harvest in northern France.
Evaluation of selectively bred eastern oyster (Crassostrea virginica) strains in Delaware Bay: implications for living shoreline enhancement
(University of Delaware, 2025) Hudock, Rileigh Elizabeth
Oyster-based restoration projects, particularly living shorelines, are being installed to protect coastal ecosystems and infrastructure. While these installations often successfully create aquatic habitats, further refinement in optimizing the growth potential of shellfish on shoreline installations will increase the success rate and efficiency of restoration projects. This study evaluated the growth and mortality of two farmed strains of oyster, NEH® (high salinity tolerant) and DBX (medium and low salinity tolerant), in the lower Delaware Bay. From July to October 2023, we monitored the growth and mortality of each strain cultured in rack-and-bag oyster aquaculture gear positioned at intertidal and subtidal environments. The effect of tidal position, temperature, salinity, and dissolved oxygen on oyster performance (average oyster length and weekly oyster mortality) was examined using Generalized Additive Models (GAMs) to determine covariate importance towards the growth and survival of both oyster strains. Our findings indicate that NEH® oysters exhibited better meat condition than DBX oysters (two-way ANOVA, p = 0.03), as well as a significant difference in average weekly length (Growth GAM, p = 0.02). Oyster weekly average length was influenced by average temperature and previous week’s minimum dissolved oxygen, while weekly mortality was influenced by minimum and average salinity and minimum dissolved oxygen. Subtidal oysters exhibited greater growth (Growth GAM, p < 0.001) and survival (Mortality GAM, p = 0.01; Scheirer-Ray-Hare test, p < 0.001; Wilcoxon rank-sum test, p < 0.001, adjusted with Bonferroni corrections) than intertidal oysters, suggesting the addition of subtidal design features can enhance the likelihood of success for oyster restoration projects. These results emphasize the need to select an appropriate strain based on local environmental conditions and suggest that pre-seeding selectively bred oysters into living shoreline materials is a viable option to enhance restoration efficiency. Our results aid in our understanding of identifying important physical and environmental factors that determine oyster performance and provide insights via statistical models that can be applied to inform restoration and shellfish-based living shoreline planning.
Observational evidence of along-shore exchange between Bransfield Strait and the central West Antarctic Peninsula
(University of Delaware, 2025) Gessay, Jakob
Melting of Antarctic continental ice has a significant impact on rising global sea levels. Glacial retreat via basal melting is strongly modulated by ocean thermal forcing from shelf bottom waters. The temperature distribution of these shelf bottom waters helps categorize shelf regions around Antarctica as either warm or cold shelves. Considerable work has been undertaken in understanding the processes that set these independent thermal shelf regimes in isolation. However, less is known about along-shore exchange between warm and cold shelves around Antarctica. Using a novel mooring array deployed from December 2022 to December 2023 I investigate the exchange between the warm central West Antarctic Peninsula (WAP) shelf and the cold Bransfield Strait shelf. Using this data, I characterize the subtidal mean flow and its (seasonal) variability with depth through key pathways into Bransfield Strait, from the south through the Low-Hoseason Island Gap, and from offshore through Boyd Strait. In general, strong subtidal mean inflows in both the Low-Hoseason Island Gap and Boyd Strait are met with weak outflowing countercurrents, however there is a notable seasonality to these flow intensities. I supplement the subtidal mean flow characterization with a tidal analysis in which I find both significant semi-diurnal and diurnal tidal motions with small ellipticities often aligning with local bathymetry. I quantify exchange by calculating volume flux into Bransfield Strait and by conducting a water mass analysis at every mooring. Flux results through both Boyd Strait and the Low-Hoseason Island Gap show a reduction of flow into Bransfield Strait during winter. The weaker flows transporting water out of Bransfield Strait is more persistent, showing little seasonality. Significant seasonal variability is also evident in the results of the water mass analysis. Warm water masses characteristic of the central WAP dominate in southern Bransfield Strait in fall months with cool water masses characteristic of Bransfield Strait peaking in spring months. To understand the frequency and intensity of cold water intrusions in the central WAP I conduct an event-based analysis at a mooring off Anvers Island showing synoptic-scale deep cooling events consistent with the seasonal variability observed in Bransfield Strait. My results help confirm prior modeling work which had yet to be validated via observations and expands our understanding of circulation in this region and how it evolves seasonally and with depth. This work has implications for calculating future regional heat budgets at depth and by season. More accurate heat budgets derived from observations will help produce more accurate models that will increase our understanding of the magnitude and variability (seasonal or inter-annual) of glacial melt and hence sea level rise.
The characterization and device fabrication of germanium-tin and titanium dioxide heterojunction
(University of Delaware, 2025) Yu, Tianqi
GeSn alloy, a group-IV semiconductor known for its tunable bandgap, offers promising performance in near- and mid-infrared optoelectronic applications. However, its low thermal stability and doping limitations also lead to challenges in device fabrication. To address the issue, Titanium dioxide (TiO2), with its reasonable chemical stability, capability for low-temperature deposition, and optical transparency in visible and near-infrared light, serves as a viable n-type component for GeSn alloy. ☐ In this research, a Heterojunction photodiode incorporating p-type GeSn and n-type TiO2 is fabricated to investigate the performance of GeSn and TiO2 incorporation. X-ray diffraction (XRD) is used to analyze the structural characteristics of various materials, while reciprocal space mapping (RSM) is employed to investigate the properties of GeSn films. Electrical measurements, such as current–voltage (I–V) analysis and CTLM testing, are conducted to evaluate device performance and contact resistance.
Coherently distributed RF antenna arrays using photonic links
(University of Delaware, 2025) Chowdhury, Shadia Islam
This thesis presents a coherently distributed multiple-input multiple-output (MIMO) radio-frequency (RF)-photonic system operating at millimeter wave (mmWave) frequencies. Over long distances with negligible propagation loss, this system uses radio-frequency-over-fiber (RFoF), or RF-photonic links, to implement distributed antenna system that preserves both the spatial and temporal coherence of RF signals. It utilizes the principles of photonic signal generation and the distribution of RF signals using a centralized optical local oscillator (OLO) to maintain phase and frequency coherence across antennas separated by tens of kilometers. Based on the proposed RF-photonic system architecture, this thesis presents the development and simulation of various antenna configurations, including single-array architectures, unconstrained distributions of subarrays and individual elements. These configurations are analyzed as proof-of-concept demonstrations to validate the precoding strategies, including zero forcing (ZF) and maximal ratio transmission (MRT), and their performance is evaluated using a simulation precoder design algorithm to establish the feasibility of the overall system design. ☐ This RF-photonic system also includes a photonic processing unit (PPU) that contains a tunable optical paired source (TOPS), photonic feed network that enables a complex weighting matrix, which enables phase and amplitude control in the optical domain to apply beamforming weights, and finally high-power photodiodes (PDs) in the remote RF frontend. The system is used to implement optimized complex weights, on distributed antenna elements, to realize “flat,” i.e., lying in the plane of propagation, holographic field profiles. In order to implement the practical performance of the holographic field profiles in the phase-synchronized distributed beamforming systems, hardware and software development are carried out. This distributed beamforming system includes the implementation of the transmitter (Tx) module, including both the software framework developed in LabVIEW and hardware enhancements achieved through the integration of supplementary test equipment for improved data acquisition and system monitoring. ☐ Various calibration techniques are used to obtain far-field radiation patterns and to maintain phase coherence across multiple distributed antennas. Experimental validation of a one-dimensional indoor system is used to demonstrate the aforementioned “in-plane” holograms, along with their comparison to simulation results. The experimental implementation and validation of various precoding schemes are utilized to implement the long range distribution of RF signals without being limited by cable loss. The proposed system preserves coherence even with differential fiber lengths to the different remote antennas. This system can be expanded to support multiple users by increasing the number of transmit antennas and incorporating data modulation on the coherent carriers. With these advancements, full system-level performance evaluation including data rate, spectral efficiency and bit error rate is planned for future work.
Implementation and analysis of FMCW radar for target detection and SAR imaging
(University of Delaware, 2025) Nair, Aiswarya
Frequency Modulated Continuous Wave (FMCW) radar systems are compact, energy efficient, and cost-effective, making them highly suitable for both commercial and defense applications. Operating with continuous low-power transmission, these systems enable accurate detection of target range by measuring the time delay between transmitted and reflected signals and target velocity from the Doppler frequency or position change over time. ☐ To increase the spatial resolution of a radar system, which depends on the antenna aperture and signal bandwidth, without adding hardware complexity, Synthetic Aperture Radar (SAR) techniques are integrated into the FMCW framework. This combination enables the development of a compact and lightweight imaging radar capable of achieving high-resolution imagery through motion. In this approach, measurements are collected at multiple spatial positions under the stop-and-go assumption, and image reconstruction is performed using the Range-Doppler and Backprojection algorithms. ☐ This thesis presents both theoretical and experimental investigations into FMCW radar signal generation, range–velocity measurement, and SAR-based imaging. The results demonstrate the effectiveness of FMCW-based SAR for multi-target localization and high-resolution imaging while maintaining a low-cost and power-efficient hardware design.
Exploring science and engineering curricula for two and three-year-olds: a multiple case study of early educators' curricula
(University of Delaware, 2025) Niranjan, Subrina D.
This study examines how the Next Generation Science Standards Content Disciplines (Life science, Earth science, Physical science, and Engineering) are addressed across two early childhood curricula. The multiple case study content analysis examines one teacher’s classroom communications to families of her two-year-old and three-year-old half-day, preschool classes for the 2022-2023 school year. Data sources included in the secondary data set are the teacher’s weekly emails to families, weekly schedules, monthly Google photo albums shared with families, and two Classroom Science Walks. These resources provide rich insight into how the early childhood educator structured her science curriculum across the academic year (2022-2023). Findings indicate that the children actively engaged in daily hands-on science explorations each day and addressed each of the core science and engineering disciplines weekly. The curriculum for each age group was designed to be developmentally appropriate (NAEYC, 2014), fitting the children's needs and allowing them to engage in science activities in both indoor and outdoor settings. Children had opportunities to learn within a supportive, play-based environment designed to connect to the children’s natural curiosity and eagerness to explore. This study highlights ways that the early science curriculum enables children to design various objects, motivates deep thinking, supports interdisciplinary learning opportunities, builds problem-solving skills, and ultimately opens doors for socialization.
Design of secure communications LIN-CP for V2G in EV charging systems
(University of Delaware, 2025) Kilaru, Go Charan
The increasing adoption of electric vehicles (EVs) has accelerated the development of Vehicle-to-Grid (V2G) systems, which enable bidirectional energy exchange between EVs and the electric grid. Reliable and secure communication between electric vehicles and electrical vehicle supply equipment (EVSE) is essential to coordinate these interactions. This thesis focuses on the LIN-CP protocol, a lightweight communication framework defined in SAE J3068 series for AC charging applications. Although LIN-CP offers a cost-effective and interoperable approach to EV-EVSE communication, it lacks security features such as encryption, integrity protection, and authentication. ☐ This thesis proposes a conceptual secure messaging extension for LIN-CP, which integrates lightweight cryptographic mechanisms such as authenticated key exchange (AKE) and authenticated encryption (AEAD). The proposed solution is designed to operate within LIN-CP’s scheduling and bandwidth constraints while preserving compatibility with existing protocol specifications. ☐ By outlining the structure and function of LIN-CP and presenting a feasible direction for protocol-level security enhancement, this work contributes to the growing need for secure, scalable, and cost-effective communication standards in V2G ecosystems.
Why do countries adopt (or abandon) gender-responsive budget policies?: the case of Australia, 2014-2022
(University of Delaware, 2025) Kim, Solbee
The term gender-responsive budgeting (GRB) encompasses a range of analytic approaches and other policy tools intended to steer public budget processes and allocations toward greater equality for men and women. By the second decade of this century, at least 80 – roughly a third – of the world's countries had adopted some version of GRB. GRB-adopting countries span the range of the world's regions and income levels, with no obvious pattern connecting region, culture, or economic conditions to the adoption or non-adoption of GRB initiatives. Why do some countries adopt GRB while others do not? As a first step toward answering that question, we used the advocacy coalition framework to structure a process-tracing examination of three recent GRB policy changes in Australia. Australia in 1983 became the first county in the world to adopt GRB and began producing an annual Women's Budget Statement (WBS) in 1984. That practice was abandoned in 2014 following a change of government but later re-adopted in 2021 by the same government in the wake of a political scandal. A new government took office in 2022 and substituted its own WBS for the previous government's several months into the 2022-2023 fiscal year. The processes leading to those policy changes were consistent with the ACF's general hypotheses about the causes of policy change but can also be explained as the results of electoral politics and partisan electoral maneuvering.
Analysis of evolutionary swarm intelligence optimization based on neighborhood topology design and benchmark functions
(University of Delaware, 2025) Rahman, Tahmidur
The evolutionary algorithms based on swarm intelligence are analyzed in this paper. The artificial bee colony and particle swarm optimization methods are described in detail to better understand the theory behind swarm optimization. The particle swarm optimization with perturbations and PSO in the software-defined network node distribution algorithm was helpful to explore the search space options more clearly. In the state-of-the-art research, while building the PSO algorithm, the balance of the objective and penalty functions is ranked stochastically, and it is tested based on 13 benchmark functions. Also, the optimum solution is dependent on the random connections of the neighborhood topology design and the relation with the benchmark functions. So, the correct optimization process algorithm building will depend on the variable selection, random connection analysis among the neighbor swarm particles, and the usage of the testing functions. In our future work, we are planning to develop an ideal algorithm to utilize all these aspects to create an environment for better network optimization.
Romantic relationships during emerging adulthood: associations with adult attachment and previous relationship experiences
(University of Delaware, 2025) Medisetti, Sanjhna
The current study explores the attachment styles and characteristics of individuals who have been in romantic relationships. This research expands previous romantic relationship experiences to include online dating, including topics of online grooming and dating app usage. Participants (n = 210) completed an online questionnaire assessing their attachment, relationship quality, childhood experiences, and online experiences. Analyses focused on examining associations among study variables. In general, results were consistent in that those participants who had insecure attachment tendencies also had more challenges in their romantic relationships, both in person and online. Specifically, insecure attachment was associated with reduced relationship quality and increased relationship difficulties, as well as with negative experiences online. Implications of these findings are discussed, and ideas for future research are explored.
Measurement of hydrogen permeability and crossover in PEM electrolyzer membranes
(University of Delaware, 2025) Anand, Srisharan
Hydrogen production via proton exchange membrane (PEM) electrolyzers is an essential technology for the transition to sustainable and carbon-free energy systems, offering efficient and scalable solutions for green hydrogen generation. However, hydrogen crossover through the electrolyzer membrane remains a critical challenge, impacting system efficiency and safety. This thesis is focused on the measurement of hydrogen permeability of novel membrane materials designed to mitigate hydrogen crossover. The membranes were then installed in a single cell electrolyzer and hydrogen crossover was measured under unpressurized and pressurized conditions. ☐ The study is divided into three phases. The first phase involves characterizing the hydrogen permeability of candidate membranes using a depressurization cell setup, where temperature and hydration level are varied to assess the membrane’s hydrogen permeability under various operating conditions. In the second phase, the selected membranes were integrated into an unpressurized PEM electrolyzer and hydrogen crossover was monitored using a gas detector. In the final phase, the membranes were tested in a pressurized PEM electrolyzer at 10-30 bar to examine performance, operational stability and hydrogen crossover under pressurized conditions. ☐ The findings of this research offer valuable insights into advanced membrane materials with significantly reduced hydrogen crossover that can enhance the performance of PEM electrolyzers under both standard and pressurized conditions. This work contributes to the development of more efficient and scalable electrolyzers for large-scale hydrogen production, supporting the global transition to a robust hydrogen economy.
Coherent uplink and downlink communications in RF-photonic distributed MIMO systems for 6G mmWave
(University of Delaware, 2025) Ullah, Md Saheed
This thesis presents multi-user beamspace processing techniques for both uplink and downlink in a distributed multiple-input multiple-output (MIMO) RF-photonic system operating at millimeter wave (mmWave) frequencies. For the uplink, a novel Beamspace Interference Rejection Combining (BIRC) algorithm is proposed to address the limitations of conventional interference rejection combining (IRC) techniques in dense electromagnetic (EM) environments. Traditional element-space IRC suffers performance degradation as the array size increases and the element-wise signal-to-noise ratio (SNR) during channel state information (CSI) estimation decreases. The BIRC algorithm overcomes these challenges by performing both CSI estimation and interference suppression in a reduced-dimensional beamspace, derived through the spatial Fourier transform and power-based beam selection, and implemented in the analog-photonic domain. This approach improves the robustness of CSI and reduces the dependence on high-speed RF chains. Simulations using the NYUSIM channel model demonstrate that BIRC outperforms its element-space counterpart, with performance gains growing with array size. ☐ For the downlink, the thesis introduces an RF-photonic MIMO architecture that enables coherent beamforming across widely distributed antennas, supporting wideband, multi-user, and multi-beam transmission. A centralized optical local oscillator (LO) is distributed to maintain phase and frequency coherence across antennas separated by tens of kilometers. Mathematical formulations establish the foundations of IQ modulation and the implementation of beamforming weights in the optical domain using phase and amplitude control. The experimental implementation and validation of various precoding schemes show strong alignment with simulation results, confirming the feasibility of coherent downlink transmission using distributed antennas connected via long-distance fronthaul. In addition, the beam squint effect is analyzed in simulation, demonstrating noticeable performance degradation when a single-frequency precoder is applied to wideband signals. ☐ A circularly symmetric cell-free (CF) MIMO system model is introduced to formulate and evaluate the system-level performance of a cell-free architecture, where multiple access points (APs) are distributed along a circular perimeter to serve users located within a concentric region of interest. A sum capacity metric is used for performance validation and is shown to closely follow the beamforming-based sum rate. Further simulation analysis demonstrates the cumulative distribution function (CDF) of the sum capacity while varying the inter-AP angle. The results demonstrate that performance improves with increased inter-AP angle due to enhanced spatial diversity. To validate this relationship, the best- and worst-performing user equipment (UE) combinations from Monte Carlo simulations are identified and visualized. Additionally, a beam selection algorithm based on volume maximization is introduced for the RF-photonic cell-free MIMO system. The proposed algorithm is compared against several existing algorithms, including the Interference-Aware Beam Selection (IA-BS) algorithm, brute-force search, maximum-power-based selection, and full-beamspace (i.e., fully digital) methods.
A computer vision pipeline for automated phenotyping of maize brace roots
(University of Delaware, 2025) Cristiano, Joseph
This thesis demonstrates a novel application of machine learning in imaging for automated phenotyping in plant science, specifically focusing on maize brace roots to enhance crop yield and resilience. It addresses the limitations of manual and semi-automated phenotyping methods for brace roots, which are labor-intensive. It highlights the challenges of accurately segmenting complex field images due to inconsistent lighting and obstructions. The work proposes a new machine learning-based approach to efficiently create datasets and develop robust models for comprehensive phenotypic analysis in diverse field conditions. ☐ A survey of related works provides foundational context by reviewing modern image segmentation techniques, including training processes, various loss functions, and optimization algorithms. The discussion differentiates between Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs), emphasizing their respective strengths in image processing and the benefits of encoder-decoder architectures like U-net for segmentation. Crucially, the role of large foundational models such as SAM2 for accelerating annotation tasks in plant science is detailed. The limitations of these models on out-of-distribution data are also acknowledged, underscoring the necessity of data augmentation techniques to enhance model robustness and generalization from limited and diverse datasets. ☐ The proposed approach outlines a comprehensive data collection strategy utilizing robotic platforms (Brobot 2021, 2023) and a handheld device (POGO stick) across multiple field seasons (2021-2024), yielding a large and varied image dataset. A novel dataset creation pipeline leverages state-of-the-art promptable segmentation models (SAM2) to efficiently annotate critical features like roots, stalks, and scale markers, significantly reducing human annotation time. To bolster model robustness against field variability, a rigorous preprocessing and data augmentation regimen, including custom padding and geometric transformations, is applied. The core segmentation architecture is a Unet with a pre-trained ResNet-34 encoder, optimized with weighted cross-entropy loss to manage class imbalance, and a classifier head is integrated to filter usable samples by leveraging latent space embeddings. Additionally, the thesis details methods for geometric analysis of segmented masks to quantify specific brace root phenotypic traits. ☐ The results section demonstrates that variability in data quality across field seasons significantly affects model performance, highlighting the critical contributions of data from 2021 and 2024 to generalization. The combination of transfer learning and data augmentation proved instrumental in achieving fast and effective convergence, with the best-performing segmentation model exhibiting strong class-wise metrics (IoU, F1 score, precision, recall) across background, markers, roots, and stalks. For classification of usable samples, a CLIP-based model achieved superior accuracy (96.95\%) compared to the custom CNN model, which was subsequently used for large-scale data filtering. Furthermore, significant positive correlations were observed between automated measurements and human-driven RootTaggingGUI measurements for key traits such as root count, spread width, stalk width, and root height, validating the efficacy of the automated phenotyping pipeline for these traits. ☐ In conclusion, this thesis successfully establishes an automated machine learning pipeline for robust and efficient brace root phenotyping in maize, effectively addressing the challenges posed by variable field imaging conditions and the need for multi-class segmentation. The methodology, encompassing advanced annotation techniques, comprehensive data augmentation, and a high-performing segmentation and classification model, offers a significant stride towards accelerating plant science research and crop improvement efforts by enabling high-throughput measurement of crucial phenotypic traits. Future work will aim to refine approximations for traits with weaker correlations and ways that segmentation can continue to improve.
Resolving lipid-nanoparticle bleb structures from small angle scattering measurements
(University of Delaware, 2025) Pathan, Afnan
Lipid nanoparticles (LNPs) are delivery vehicles for nucleic acid payloads, such as siRNA and mRNA. Typical LNPs consist of the ionizable lipid, cholesterol, phospholipid, PEG-lipid, and nucleic acid, self-assembling into a nanostructure for biodistribution, endocytosis and endosomal escape necessary for RNA delivery and transfection. Recent studies report protruding structures on otherwise spherical LNPs, forming a two-compartment lipid-rich and aqueous-rich morphology termed LNP “blebs”. While Cryo-TEM measurements can identify bleb architectures, such measurements are limited in statistical sample size and require delicate sample preparation. Small-Angle Scattering (SAS) measurements of LNPs in formulation provide an alternative route to identify blebs, but existing analytical models are limited in resolving structural features. In this work, we present a computational workflow for efficient analysis of SAS experiments of LNP blebs to resolve their microstructure and composition. This workflow numerically calculates the pair distance distribution function (PDDF) for model LNP bleb structures via a simple Monte Carlo method, weights it by the product of excess scattering lengths based on the composition, and generates simulated SAS profiles by performing a 1D Fourier transform. ☐ A population-based evolutionary optimization algorithm, differential evolution (DE) is used to minimize the discrepancies between the simulated and experimental scattering profiles, refining structural parameters to reconstruct the representative morphology for the experimental sample. Recently published small-angle neutron scattering (SANS) experimental data of LNP-self-amplifying (SAM) RNA formulations (Thelen et al., ACS Nano, 2024) are analyzed. ☐ Simultaneous analysis of multiple SANS profiles measured under different solvent compositions enabled the quantification of average-size, structure and distribution of lipid components and RNA in the LNP bleb representative morphology for the formulations. Our computational workflow provides an efficient, automated, and physics-based approach for SAS analysis of complex morphologies and is shown to be an improvement over the prior, analytical approximations. The LNP model structure development and efficient calculation of PDDF shown in computational workflow is used for generating high-throughput data set of structure and PDDFs for a library of LNP morphologies. The data set developed is utilized in a separate work for developing Physics Informed Machine Learning model for LNP structure prediction. Additionally, a two-compartment core-shell bicelle analytical model is developed and tested as an approximate model for SAS analysis of LNP blebs which can be easily integrated with widely used SAS analysis software. The insights into the structure and composition of LNP blebs aid in the rational design of LNP formulations.
Phase separation dynamics of ternary system with and without shear: a Lattice Boltzmann Method approach
(University of Delaware, 2025) Arumugam Kumar, Gokul Raman
Ternary phase separation plays a crucial role in understanding the complex behavior of multi-component fluid mixtures, particularly under varying flow conditions. This study investigates the phase separation dynamics of ternary fluid mixtures under both quiescent and sheared conditions using a free-energy Lattice Boltzmann Method. In the absence of shear, domain growth and final morphologies are characterized for different volume fraction configurations across a range of values of the intrinsic fluidity parameter, defined as the ratio between peclet and capillary number. ☐ When a ternary mixture subjected to shear flow undergoes phase separation, the resulting morphology is governed by two competing effects: the natural coarsening of domains and shear-induced deformation. While domains tend to grow over time, the applied shear stretches them along the flow direction, leading to unique morphologies. Our results for an intermediate shear regime (applied shear is insufficient to induce strongly anisotropic or fully aligned structures), reveal that at low capillary numbers, the system reaches a periodic steady state featuring complex droplet morphologies such as double emulsions and worm-like structures. In contrast, at high capillary numbers, phase separation results in banded structures extended along the shear direction. Under inertialess conditions, we find that the phase separation dynamics are primarily governed by capillary number, and appear to be largely independent of the components' volume fractions. This contrasts with the no-shear case, where the final morphologies are strongly dependent on the volume fractions. ☐ The range of morphologies predicted under both quiescent and sheared conditions demonstrates the potential of this research for applications in the design and manufacturing of polymeric and soft materials.
Characterization of groundwater dissolved organic carbon along the transition zone from saltmarsh to upland in the Delmarva Peninsula: influences of redox conditions and hydrological processes
(University of Delaware, 2025) Sprague-Getsy, Amanda
In the face of sea level rise (SLR), salt marshes are migrating inland creating a transition zone between the marsh and upland. Little is understood about how the critical ecosystem services and biogeochemical activity in this zone are evolving in response. This research aims to characterize the organic carbon in shallow groundwater along the transition from salt marsh to upland and identify the major drivers of biogeochemical dynamics; specifically, how hydrological processes impact redox conditions and dissolved organic carbon (DOC) cycling. Groundwater was extracted and in-situ redox probes installed along transects across the marsh-upland transition zone at agricultural and forested sites in Delaware, Maryland, and Virginia, USA. DOC concentration and composition (SUVA254 and PARAFAC component modeling) were characterized and related to groundwater level and redox conditions across storms, tidal cycles, and seasons. DOC concentrations did not significantly change during Hurricane Ian (October 2022), but the SUVA254 and carbon components did change throughout. Similarly, while DOC concentrations remained stable over a tidal cycle, DOC composition varied. Samples collected during the summer showed low DOC concentrations in comparison to samples collected in spring and winter, possibly influenced by seasonal decline in the water table. Land cover also influences DOC concentrations, which were higher in forested than agricultural sites. Results from this study can be applied to better predict how groundwater carbon dynamics along the marsh-upland transition will respond to hydrologic and climatic shifts.
Defining forming limits of highly aligned discontinious fiber composites
(University of Delaware, 2025) Morris, Kyle S.
Carbon fiber reinforced composites have become a material of interest in various high-performance sectors such as aerospace and automotive applications due to their high stiffness to strength ratio, allowing for improved durability and energy saving. Typically, carbon fiber reinforced composites are made with continuous carbon fiber which imposes limitations on manufacturability due to inextensibility of the fibers. Highly aligned discontinuous fiber (ADF) composites have been shown to achieve aerospace-grade properties and have the additional advantage to stretch form biaxially to complex geometries. This thesis focuses on evaluating the formability of ADF composites utilizing Tailored Universal Feedstock for Forming (TuFF) aligned fibers combined with thermoplastic and thermoset resin systems. The primary aim is to establish a comprehensive process-structure-property relationships for ADF composites through novel methodologies for defining forming limits and characterizing material performance. ☐ A novel aligned discontinuous fiber forming limit diagram (ADF-FLD) was developed to define formability for this class of material relative to material orientation. Methodologies were developed to construct a forming limit diagram (FLD) for ADF composites, providing a detailed framework for strain mode forming limits based on lamina fiber orientation and a predicted thickness variability in a closed mold and an open mold forming process. To demonstrate this, ADF composite blanks were stretch formed to various strain levels and modes (longitudinal plane strain, transverse plane strain, and biaxial plane strain), while employing both in situ and ex situ techniques to measure deformation. By manipulating surface ply orientations of ADF laminates (0, 45, or 90 degrees) relative to the major strain direction, different strain modes were imposed and measured using photogrammetry post forming and digital image correlation (DIC) for real time strain analysis. ☐ First, a method was developed to characterize deformation of multiaxial thermoplastic TuFF laminates in a series of closed molds at various strain levels, using a double diaphragm gas bulge forming process and photogrammetry to analyze strains post forming. A first-order failure definition based on a predicted thickness coefficient of variation relative to average strain was employed to evaluate the forming limits of the material and populate an ADF-FLD, describing the formability of the ADF composite. ☐ Additionally, a method was devised to characterize the deformation response of multiaxial thermoset TuFF laminates in diaphragm forming for longitudinal and transverse plane strain. Using a gas bulge method and an Interlaken SP75 highly instrumented forming press, high fidelity strain measurements were obtained via an in situ 3D digital image correlation system. This method allowed for real-time recording of surface strains, allowing for progression of strain variability to be measured continuously. The same failure criterion was applied to the strain data, and an ADFFLD was constructed with repeatable results for each strain mode. ☐ Overall, this thesis provides a comprehensive methodology and experimental framework for defining forming limits and optimizing the stretch forming process of ADF composites. The outcomes offer significant insights into the material and process variables, crucial for the design and manufacturing of complex composite parts in aerospace and other high-performance applications.
"I have no idea really what's going on": teachers' self-efficacy and executive function implementation during the COVID-19 recovery year
(University of Delaware, 2025) Mykyta-Chomsky, Rosa
This thesis examines teacher perceptions of their self-efficacy in implementing executive function instruction during the 2020–2021 school year, described here as the COVID-19 recovery year. Executive function skills—including inhibitory control, emotional control, working memory, cognitive flexibility, and organizational skills— are often supported through consistent classroom routines. The disruptions caused by the pandemic required teachers to rebuild these routines across in-person and virtual settings. These disruptions resulted in challenges that could potentially influence teacher self-efficacy. Using teacher self-rating and qualitative interviews, this study explored how learning environments influenced teacher perceptions of self-efficacy in supporting student executive function development. Findings suggest that the classroom setting and teacher perceptions of self-efficacy influenced executive function instruction, particularly in the domains of emotional and inhibitory control. Student-teacher relationships emerged in the qualitative data as a key support for teacher self-efficacy, aligning with existing literature (Boe et al., 2008; Sharpe & Strosnider, 2022). These results indicate that, beyond instructional mode, multiple factors shape teacher self-efficacy in executive function instruction, extending the relevance of these findings beyond the COVID-19 context.
Impacts of water and residue management on soil health indicators in rice paddy soil
(University of Delaware, 2025) Rosado, Baileigh
Soil health is the foundation for successful agriculture, and indicative tests have been developed to assess soil health for soils that support many upland crops. Healthy soil is typically characterized by key biological, chemical, and physical traits, including an active microbial community, high organic carbon, and stable aggregation. Rice, a globally essential food crop, presents unique soil health challenges due to its cultivation in flooded conditions, and it is unclear if traditional soil health frameworks apply to rice cultivation. Key concerns in rice agroecosystems include elevated grain As levels, excessive irrigation water use, and CH4 emissions, which threaten human health, environmental sustainability, and contribute to climate change. This study explores how water and residue management strategies, factors that heavily influence As and C cycling, affect soil health test results. We found that, of the 26 soil health indicators tested, 15 were sensitive to water management and 20 were sensitive to residue management, even though water had a stronger impact on grain As concentration and CH4 emissions than crop residues. Comprehensive Assessment of Soil Health (CASH) scoring, modified to include grain As and CH4 emissions, indicated that non-flooded and straw-amended paddies produced the healthiest soil. Of the 26 soil health tests used, 16 had significant correlations with yield, 11 with grain As, and 18 with CH4 emissions. Mehlich-3 extractable Fe was highly affected by water and residue management, and had the strongest correlations with yield, grain As, and CH4 emissions. By identifying soil health tests that are sensitive to these management practices and addressing As and CH4 concerns without compromising yield, this research aims to provide rice farmers and researchers with practical tools to enhance soil health while balancing environmental and human health considerations.

Browse

Recent Submissions