-
Prospective Students
-
Faculty and Staff
-
Current Students
-
Alumni and Friends
Gas Metal Arc Welding (GMAW) was investigated as a method for the rapid Wire Arc Additive Manufacturing (WAAM) of magnesium alloys. Magnesium AZ61a deposition wire was used to build multilayer walls, large blocks, and hollow cylinders using both high and low input-energy-rate (IER) parameters. The printed structures were analyzed to determine mechanical properties, microstructure, and porosity. Multilayer-wall samples printed at the same torch travel speed (TTS) showed a material yield strength (YS) of 120 MPa, independent of print orientation in relation to the applied tensile test pull force. The samples that were printed at a faster TTS showed the same response to loading conditions, but had a lower YS of 106 MPa, thus demonstrating how an increase in TTS lowers the YS of the deposited material. The stress-at-fracture for all these samples was between 260 MPa and 270 MPa. For the large multi-layer/multi-row (MRML) samples a YS of 120 MPa was also obtained but with lower stress-at-fracture points between 150 MPa and 220 MPa depending on print orientation, due to the presence of larger internal defects caused by bead overlap issues. Scanning Electron Microscope (SEM) analysis was performed on the fracture surface, showing ductile behavior in the fused regions and also uncovering material defects in the MRML samples such as trapped spatter, trapped gas bubbles, and cracks. Optical micrographs were obtained to analyze the microstructure of the samples in the heat affected zone (HAZ) as well as in the bulk material. Grain refinement from 38 µm pre-weld down to 10 µm and 28 µm post-weld was determined for MRML blocks and multilayer walls, respectively. Multilayer hollow cylinders were printed to test the ability of the method to produce closed-shape parts. These cylinders were produced at both high and low IERs and yielded parts with post-deposition machined wall thicknesses ranging from 1.5 mm to 4.5 mm. X-ray Computed Tomography (XCT) was performed to determine the porosity of these parts. The three sections analyzed showed a total-part percent porosity of 0.04 %, 0.039 %, and 0.07%. Larger individual defects, particularly at the closure-of-bead zone were detected, with a maximum single layer percent porosity of 0.8 %. Lastly, a Finite Element Analysis (FEA) model was created to simulate the deposition of the beads and the heat transfer throughout the process. The element activation feature in COMSOL Multiphysics was combined with the simulated torch path to model the deposition of the material. Heat transfer modes of conduction, radiation, and convection were conditionally assigned to the boundaries of the substrate and of the beads as functions of time and material deposition. The Goldak double-ellipsoid heat source was used as the input heating method for the substrate. To simulate the true-to-life GMAW process, where already molten material drops onto the substrate, a bead pre-heating function was created and applied to the inactive elements of the bead before it gets deposited during the simulation so that when the elements are activated above the weld pool they are at the correct temperature.
Traditional cyber defense approaches lack the necessary agility to effectively counter stealthy and undetectable attacks, placing defenders at a disadvantage. In response, Active Cyber Deception (ACD) has emerged as a promising solution by dynamically orchestrating deceptive environments to mislead and corrupt attackers' decision-making processes. However, the development of efficient and effective deception systems requires the integration of human intelligence and comprehensive malware analysis to understand attack behaviors and automate deception strategies.
This dissertation presents three innovative approaches in the field of ACD. Firstly, DodgeTron combines dynamic analysis using symbolic execution tools and machine learning to automate the creation of deception schemes against malware by categorizing malware into known families and utilizing HoneyThings. Secondly, symbSODA performs dynamic analysis on real-world malware and data flow analysis to extract malicious sub-graphs (MSGs) and map them to the MITRE ATT&CK framework using Natural Language Processing. This enables the creation of a Deception Playbook for deceiving specific malicious behaviors with deceptive API hookings. Finally, ranDecepter integrates active cyber deception to identify ransomware in its early stages and employs binary orchestration methods to repurpose the malware as a channel for exhaustively transmitting encryption information (including keys) to the attacker, effectively depleting their available resources.
Comprehensive evaluations validate the accuracy and effectiveness of these approaches in deceiving adversaries, reducing analysis time, and mitigating malware threats. This research significantly contributes to the field of active cyber deception and offers efficient and scalable solutions for protecting digital systems against sophisticated adversaries.
Dense and porous Silicon Carbide (SiC) ceramics and composites are used in a wide range of applications that require high thermal, mechanical, and electrical stability, excellent corrosion, and wear resistance. However, manufacturing of SiC through conventional powder metallurgy technique techniques is often challenging due. Due to the covalent bonding between Si and C, they have a high melting point. Hence high temperatures, pressures, and controlled atmospheres are required during sintering to manufacture SiC ceramic with good mechanical and thermal strength. Other techniques to manufacture SiC at relatively low temperatures involve thermal oxidation, pressureless sintering, and the addition of sintering additives. Some applications like biological scaffolds, ballistic armor, space mirror substrates, and ceramic filters involve complex geometries. Manufacturing of complex geometries through the conventional route involves machining or molding. Machining SiC is a challenge due to its extreme hardness and abrasiveness. Molding a pre-form utilizes polymer resin which can cause shrinkage to the final product. upon debinding and sintering. Hence, the additive manufacturing route is considered feasible for the manufacturing of SiC ceramics or composites. Additive manufacturing (AM) enables 3D printing of complex geometries from a CAD model. Multiple direct AM methods were realized for the printing of SiC such as selective laser sintering (SLS), selective laser melting (SLM), stereolithography (SL), direct ink writing (DIW), and binder jetting (BJ). Among these techniques, the binder jetting technique was found to be easier to manufacture complex geometries of SiC as it does not require, i) polymer additives that cause shrinkage of the part upon sintering and it doesn’t require, ii) high laser power to melt SiC, and iii) ceramic slurry, where the amount of ceramic used is less. Binder jetting also allows the mixing of different ceramic particles and additives that can help in the densification and strengthening of the printed part. In this work, the following areas are addressed: i) a route for densifying and strengthening the powder bed binder jet-printed SiC through the mixing of particles of different sizes, formation of siloxane bonding, secondary surface modification, and sintering, ii) strengthening and densification of SiC composites using mineral binders using powder metallurgy technique, and iii) realizing the properties of SiC-mineral binder composites for space mirror and thermal applications. Part (ii) of the project was preliminary work done in order to realize the outcomes of SiC-mineral binder composites in strengthening so that it can be adopted into additive manufacturing mentioned in part (i). Future work will involve the inclusion of SiC-mineral binders into the feedstock in a powder bed binder jet in order to reduce the voids between the interspace of SiC particles and to have a strong interfacial region comprising of mineral binders that can fuse the SiC particles together and densify the printed part. This eliminates the need for the post-processing techniques such as melt infiltration, polymer impregnation, or chemical vapor infiltration.
SiC ceramics are 3D printed into cylindrical discs in a powder bed binder jet using a water binder. In this method, SiC of an average particle size of 40 µm was surface activated with NaOH to form a silica gel layer at room temperature, to which, 30% of 2 µm and 600 nm SiC particles were added and mixed homogeneously through milling. The presence of OH- ions in silica gel, creates a repulsion between SiC particles which eliminates agglomeration of particles upon spreading. The mixing of coarse and fine particle sizes reduced the percentage porosity by 50%. The as-printed green part was heat treated to 650 °C for 5h to create siloxane bonding which provided an improved handling strength. The heat-treated parts were then impregnated in various concentrations of NaOH to create silica gel through secondary surface activation. SEM images showed that the impregnated samples had more silica nucleation droplets that gave rise to silica nanowires upon sintering at temperatures between 800 °C – 1000 °C. The silica nanowires are responsible for fusing the SiC particles and bridging the pores. The optimum NaOH concentration for secondary surface activation, sintering temperature, and dwell time were determined. A 100% increase in the strength of the SiC was obtained in the samples heat treated at 650 °C, impregnated in 20% NaOH, and sintered at 1000 °C for 24h. Moreover, the formation of nanowires under an oxygen environment proved that silica nanowires can be formed at a temperature as low as 800 °C, and in air, the discs are oxygen deprived which hinders the growth of silica nanowires. Hence the mechanism of the growth of nanowires was found to be similar to the solid-vapor phase deposition. Cordierite and spodumene are silicate minerals that are known for their excellent thermal properties namely nearly zero thermal expansion coefficient. SiC is a ceramic with excellent mechanical and thermal properties. SiC, Cordierite, and Spodumene are materials that are considered for space mirrors, mirror substrates, and high-temperature applications. However, the glass ceramic form of Cordierite and Spodumene are less considered for space applications due to their poor stiffness and fracture toughness. On the other hand, SiC is highly considered for such applications however manufacturing them is a challenge considering their high melting point and hardness. Hence, in this work, a combination of SiC and the mineral format of cordierite and spodumene is introduced as SiC-mineral binder composites. SiC-mineral binder composites are 80% SiC and 20% Cordierite or Spodumene minerals prepared through the powder metallurgy technique. SiC-mineral binder composites were found to have good mechanical and thermal properties and can be a promising candidate for space mirror applications. SiC-mineral binders were combined with 1% of NaOH, pressed at 250 MPa, and heat treated to 1200 °C for 8 h. The SEM-EDX analysis showed a strong interfacial region of cordierite or spodumene fusing the SiC particles together. The fracture mechanism was found to be transgranular which is due to the strong interfacial bond that was created by the atomic diffusion of Si and Al at the grain boundary of SiC and the mineral interface. The characterization involves the comparison of SiC-mineral binders to the control SiC-cristobalite without mineral binders. The phase analysis from XRD showed the presence of cordierite, spodumene, and cristobalite phases. A transformation of β-SiC to α-SiC was also observed. A slight shift in the d-spacing, the lattice constants, and crystallite size was observed as a result of a solid solution of phases. The density and porosity of these composites were measured using Archimedes and mercury porosimetry. Further pore size analyses were done using SEM and ImageJ analysis. The results showed that the introduction of mineral binders reduced the pore size and the porosity percentage. The compressive strength of the SiC-Cordierite and SiC-Spodumene was 282.57 MPa and 184.58 MPa which was much higher than the control SiC-Cris, 97.45 MPa. The average compressive strength of SiC-Cord was three times higher than control SiC-Cris (p < 9.7 x 10-7) and two times higher than that of SiC-Spod (p <0.003). Moreover, the average compressive strength of the SiC-Spod was significantly higher than that of the control SiC-Cris (p <9.8 x10-7). Elastic modulus was found using the nanoindentation technique and was 380.54 GPa and 341.04 GPa for SiC-Cord and SiC-Spod composites. Thermal shock resistance is an important factor for materials to qualify for space applications. SiC-mineral composites showed excellent thermal shock resistance and dimensional stability when quenched from 1200 °C to room temperature. A thermal expansion coefficient of 3 x 10-6 /K was obtained for both SiC-Cord and SiC-Spod composites. The SiC-mineral composites were polished to a mirror finish and the surface roughness of areas comprised of SiC particles along with the mineral binder without pores measured using atomic force microscopy was 20.89 nm. The mean roughness of the SiC microconstituent in the SiC-Cord composite was found to be 2.37 ± 0.28 nm. Owing to these excellent thermos-mechanical properties, SiC-mineral binder composites are promising candidates for space mirror applications, mirror substrates, substrates for high-temperature devices, and catalytic converters. Also, porous SiC-mineral binder composites can be used as gas/fuel filters for automobile industries.
Recent advances in the field of Natural Language Processing, specifically in Natural Language Generation (NLG) towards Dialogue Systems have focused mainly on two-party conversations. However, group conversations or multi-party conversations (MPC) are just as prevalent in our everyday lives. While the area of multi-party conversation modeling has received some attention in recent times, MPC lacks resources for 1) corpora in differing settings (formal/informal, synchronous/asynchronous), 2) dialogue models which can participate in informal open-domain settings while maintaining speaker information, and 3) evaluation metrics which provide better insights into the performance of MPC models when it comes to operating in groups and interacting with multiple participants. We thus take a three-pronged approach towards contributing to research in the MPC modeling research area. For corpora collection, we contribute a mock social media tool that can be utilized for collecting asynchronous MPC conversations called Community Connect and utilize it for 3 experiments to collect everyday talk. For MPC modeling, we propose a response generation model, using large language models (LLMs) and graph structured networks, which is capable of taking participant relations into account towards maintaining multiple persona profiles and generating responses keeping the speaker characteristics in mind and responding accordingly. Lastly, for MPC evaluation, we present an expansion to the taxonomy of errors which specifically contributes MPC-specific metrics to the overall NLG errors. In addition to the taxonomy, we contribute to better evaluation standards across which progress in the tasks within MPC can be tracked more saliently. Through these contributions, we aim to fill the necessary gaps towards advancing MPC understanding and modeling, while also providing the tools to gauge progress until now.
Plant-specialized metabolites play pivotal roles in adapting to dynamic environments and promoting human health. Among these metabolites, glyceollins and soyasaponins hold particular importance in responding to environmental stresses and contributing to sustainable human nutrition, including the development of novel pharmaceuticals. Glyceollins are phytoalexins induced in legume species, derived from the isoflavonoid branch of the phenylpropanoid pathway, while soyasaponins belong to the triterpenoid class and are naturally abundant in legume species without requiring induction. Despite their significance, the genetic basis underlying glyceollin induction and soyasaponin biosynthesis remains poorly understood. Furthermore, previous studies on their genetic basis have primarily focused on model or major crop species, with limited research on wild crop species, such as wild soybean. To address these knowledge gaps, our study utilized wild soybeans, known for their abundant and unexplored genetic diversity compared to cultivated soybeans, to unravel the genetic basis of glyceollin induction and soyasaponin production. We employed a metabolite-based genome-wide association (mGWA) approach and identified eight SNPs on chromosomes 3, 9, 13, 15, and 20 significantly associated with glyceollin induction. Six genes near a significant SNP (ss715603454) on chromosome 9 formed two clusters, encoding enzymes of the glycosyltransferase class. Furthermore, we discovered transcription factor genes, such as MYB and WRKY, within the linkage disequilibrium of the significant SNPs on chromosome 9. Epistasis and strong selection signals were also detected for the four significant SNPs on chromosome 9, indicating their role as major evolutionary factors influencing glyceollin variation in natural populations. Moreover, to investigate the genetic basis of phytochemical diversity, we conducted comprehensive phenotyping using LC-MS analysis on an association panel of 190 wild soybean ecotypes from diverse natural environments. Among the 874 metabolite peaks detected, we successfully annotated 485 metabolites. We identified 1155 SNPs significantly associated with 359 metabolites by performing a genome-wide association study. Clustering analysis revealed eight QTLs, named QTL-multiple metabolite clusters, showing significant associations with identified metabolites. By mining data within the linkage disequilibrium blocks encompassing these QTLs, we identified 612 annotated genes. From this set, we selected 16 candidate genes based on their relevance to the triterpenoid and phenylpropanoid-derived isoflavonoid biosynthetic pathways. Among these 16 candidate genes, UDP-dependent glycosyltransferase (UGT) was considered a promising candidate gene. Sequence analysis of this UGT gene in 46 wild soybean ecotypes unveiled two haplotypes with three SNPs on exon-1, leading to amino acid changes. The following association analysis showed these two haplotypes were significantly associated with high and low soyasaponin production. These two haplotypes also exhibited notable differences in expression levels. Our findings contribute valuable insights into the genetic mechanisms underlying phytochemical diversity, specifically the induction of glyceollins and the production of soyasaponins. This knowledge is instrumental in developing climate-resilient, high-value crops with enhanced medicinal properties, ultimately benefiting both plant and human health.
This dissertation describes three studies examining how people with developmental disabilities (DD) experience social capital from childhood to young adulthood. The first study follows PRISMA standards for scoping literature review to investigate social capital in DD research from childhood to emerging adulthood. Results describe social capital definition, measurement and application in DD research and identify gaps in the literature. The second study describes extracurricular activity (EA) participation of children with and without DD and the associations of childhood EA with mental health in young adulthood. Results show: differences in EA participation of children with and without DD; EA is associated with lower psychological distress and greater flourishing among people with and without DD. The third study uses photovoice to address the meaning of interdependence for college students with DD. Participants described their experiences using photos and stories. Themes included: openness to being helped, foundational role of families, experiencing new and challenging things, and tension between wanting to help and vulnerability of being a helper. Understanding social capital in the lives of people with DD can guide policies and supports, promoting improved quality of life. This research is a steppingstone toward a more inclusive and supportive society for people with DD.
This dissertation aims to contribute to our understanding of Chinese government engagement on social media platforms. In the first section, this dissertation proposes a fresh framework for identifying persuasive techniques in textual posts. In the second section, the dissertation investigates the similarities and differences in the Chinese government's posts across various platforms, each targeting distinct audiences. The final section of this dissertation provides a quantitative analysis of users' opinions towards the content produced by the Chinese government on online social networks.
There have been an increasing number of materials developed that show multifunctional chromogenic properties (such as electrochromism, electrofluorochromism, or photochromism), but to date, few materials have shown all three properties. Materials that are electrochemically and optically active are attractive for a diverse set of applications that include smart-windows, lighting, sensing, energy production, and conservation. Achieving systems made from organic, cost-effective, readily synthesized materials would make them easy to utilize in a variety of fields. Multifunctional chromogenic dipyridinium thiazolo(5,4-d)thiazole (TTz) show promise in achieving these needs as they offer high contrast color change, high fluorescent quantum yields above 90%, and water processability while made from inexpensive starting materials. The planar, rigid, heterocyclic TTz core improves stability and reversibility as the TTz reduces from yellow TTz2+ to purple TTz•+ to blue TTz0 compared to other viologen systems.
When implemented in a low-cost poly(vinyl alcohol) (PVA)/borax hydrogel device using conductive glass electrodes, the TTz can change color and fluorescence intensity with applied voltage or light exposure. The electrochromism offers 75% transmittance contrast that is stable for 250 on/off cycles and electrofluorochromism with >90% contrast. By adjusting gel components and coating/drying the hydrogel, a variety of photochromic thin films were produced. Remarkably, the TTz-embedded films retain their high contrast chromogenic properties showing photochromism (yellow TTz2+ to blue TTz0 color change) and photofluorochromism after only one minute of light exposure. After turning blue, the oxidation back to yellow occurs through interactions with oxygen. This is potentially an effective way to optically gauge the presence of oxygen which is useful for a variety smart packaging applications for food, pharmaceuticals, and electronics. The color change speed and contrast can be tuned by adjusting borax and TTz concentrations. Because the TTz’s show multifunctional capabilities, photo-charging battery devices are tested to indicate the creation and storage of electrical charge when illuminated. When paired with the appropriate catholyte and membrane, TTz shows evidence of photocharging in hydrogel and film devices. Comparing charge discharge curves of the battery devices, illumination can increase voltages by 0.2 V and improve charging capacity. This work shows the remarkable multifunctional electroactive and photoactive properties of dipyridinium thiazolothiazole materials as well as their implemented to yield reversible, high contrast electrochromism, electrofluorochromism, photochromism, photofluorochromism, and light responsive charging.
At-home DNA testing and sharing in public genealogy databases are becoming widespread. This will facilitate finding out ancestry, genetic relatives, biological parents, making new connections, advancing medicine, and determining predisposition to various diseases and health issues. While the biomedical community glorifies the uses of the genomics revolution, the expanded obtainability of such sensitive data has substantial implications for individual privacy as genes carry sensitive personal information about human traits and predispositions to any diseases. Furthermore, DNA data has identification capability (e.g., forensics) as well as reveals familial interconnections. However, commercial DNA testing is not vigorously governed by any laws and policies. The privacy implications of public DNA data sharing remain largely unexplored. This dissertation explores users' privacy concerns and proposes a method for communicating the risks to users to inform users when sharing their DNA data.
In the first study, we explored users' perceptions regarding DNA data. We asked about their views of at-home DNA testing and sharing, followed by their expected benefits and concerns. We also talked about public genealogy databases like GEDmatch. We focused on understanding the users' preferences and perceptions on the disclosure of their genetic information under the different types of platforms and entities. Our results show that users are mostly unaware and uncomprehending of the interconnected nature of genetic data. We noted users' general perceptions and focused on understanding their preferred privacy controls while sharing their DNA data, their desired settings, policies, and rules.
From this study, we identified the need to develop a privacy-enhancing technology such that the users can make an informed choice while sharing DNA data. We also found that several policies and settings should be to preserve the privacy of sensitive data. With these findings in mind, the ultimate objective of this dissertation is to design and implement privacy risk communication methods that aid users in comprehending the risks and benefits associated with sharing DNA data, as well as enhancing transparency in access control. To evaluate the effectiveness of our developed risk communication approach, we deployed it within an existing platform, allowing us to assess users' decision-making processes and gain a deeper understanding of the nature of DNA data.
At-home DNA testing and sharing in public genealogy databases are becoming widespread. This will facilitate finding out ancestry, genetic relatives, biological parents, making new connections, advancing medicine, and determining predisposition to various diseases and health issues. While the biomedical community glorifies the uses of the genomics revolution, the expanded obtainability of such sensitive data has substantial implications for individual privacy as genes carry sensitive personal information about human traits and predispositions to any diseases. Furthermore, DNA data has identification capability (e.g., forensics) as well as reveals familial interconnections. However, commercial DNA testing is not vigorously governed by any laws and policies. The privacy implications of public DNA data sharing remain largely unexplored. This dissertation explores users' privacy concerns and proposes a method for communicating the risks to users to inform users when sharing their DNA data.
In the first study, we explored users' perceptions regarding DNA data. We asked about their views of at-home DNA testing and sharing, followed by their expected benefits and concerns. We also talked about public genealogy databases like GEDmatch. We focused on understanding the users' preferences and perceptions on the disclosure of their genetic information under the different types of platforms and entities. Our results show that users are mostly unaware and uncomprehending of the interconnected nature of genetic data. We noted users' general perceptions and focused on understanding their preferred privacy controls while sharing their DNA data, their desired settings, policies, and rules.
From this study, we identified the need to develop a privacy-enhancing technology such that the users can make an informed choice while sharing DNA data. We also found that several policies and settings should be to preserve the privacy of sensitive data. With these findings in mind, the ultimate objective of this dissertation is to design and implement privacy risk communication methods that aid users in comprehending the risks and benefits associated with sharing DNA data, as well as enhancing transparency in access control. To evaluate the effectiveness of our developed risk communication approach, we deployed it within an existing platform, allowing us to assess users' decision-making processes and gain a deeper understanding of the nature of DNA data.
