Organizational Unit:
George W. Woodruff School of Mechanical Engineering
George W. Woodruff School of Mechanical Engineering
Permanent Link
Research Organization Registry ID
Description
Previous Names
Parent Organization
Parent Organization
Organizational Unit
Includes Organization(s)
Organizational Unit
Organizational Unit
Organizational Unit
Organizational Unit
Organizational Unit
ArchiveSpace Name Record
Publication Search Results
Now showing
1 - 10 of 940
-
ItemHigh Shear Arterial Thrombosis: Microfluidic Diagnostics and Nanotherapeutics(Georgia Institute of Technology, 2019-12-20) Griffin, Michael T.Ischemic cardiovascular events remain the leading causes of death in the world, largely due to ineffective preventative therapies and diagnostic tools. This work investigated the development of a physiologically relevant, low-variability microfluidic thrombosis assay (MTA) capable of screening therapy efficacy. First, an experimental design was implemented to assess the effects of geometry, collagen surface coverage, and anticoagulant selection on MTA occlusion time (OT) variability. It was found that better control of shear rates through novel grayscale lithography techniques decreased OT variability. Fibrillar collagens was also found to have a significant impact. The MTA was then implemented to study the effects of current antiplatelet therapies, aspirin and Plavix, as compared to the endpoints of other platelet function tests (PFTs). It was found that aspirin use significantly increased MTA OT but did not prevent occlusion in the MTA. Results from Plavix use found a stronger response, where 20% of patients had complete OT inhibition. Comparison with other PFTs found that the MTA more closely matched the rates of ischemic events from larger clinical studies. Finally, the MTA was utilized to assess a nanoparticle therapy hypothesized to function through biophysical mechanisms. It was found that negatively charged nanoparticles were more effective than neutral or positively charged nanoparticles. The antithrombotic effect of charged nanoparticles persisted even with different base materials, but the effects of nanoparticle size were not consistent between materials. A mouse bleeding model was also used to show that hemostasis was maintained with the nanoparticle therapy. The implications of all results for clinical diagnostic and future antithrombotic therapy research are discussed.
-
ItemADHESION EVALUATION AND ASSEMBLY PROCESS DEVELOPMENT FOR PRINTED SILVER INK ON FLEXIBLE SUBSTRATES(Georgia Institute of Technology, 2019-11-20) Taylor, Christine L.Flexible substrates with printed electronics are being increasingly sought for the widespread and cost-effective use of flexible electronics. With printed ink on flexible substrates, several items need to be examined: synthesis of ink, deposition of ink, curing of ink, line and spacing of ink, adhesion of ink, fracture strength of ink, electrical characteristics of ink, etc. Among these items, adhesion of ink to the substrate plays an important role in the overall reliability of printed ink on flexible substrate. In this work, the adhesion and interfacial characteristics of printed conductors were determined though modified shear and peel experimental techniques. Modification to the tests were needed for handling the ink-jet printed films due to manufacturing considerations. (1) These films often are designed to be porous so that the films are more flexible by lower the stiffness. (2) Ink-jetting films often are composed of thin layers that are stacked-up to reach the desired thickness. (3) Depending on the tool and the file conversion to bitmap (or designated file extension) for the printer, the film may show indication of the path taken by the tool head with slight spaces between passes. A numerical model based on sequential crack growth was developed to examine how much the plastic deformation accounts for the experimentally measured peel energy. Lastly, a fully-additive printing process was demonstrated with resistors which resulted in around 6.5 MPa shear strength. For the assemblies, the joint strength of the ink for the joint to traces was stronger than the adhesion strength of the ink to flexible substrate.
-
ItemDevelopment of a microengineered human blood-brain barrier model with 3D astrocytic network(Georgia Institute of Technology, 2019-11-12) Ahn, Song IhThe blood-brain barrier (BBB), a unique vascular border in the central nervous system (CNS), has a highly selective barrier function that prevents unwanted substances from entering the brain. To deliver drugs into the brain, CNS delivery systems have been widely explored to cross the BBB. However, the lack of experimental models that can precisely analyze the interactions between the BBB and delivery platforms restricts successful clinical translation of CNS therapeutics. Despite valuable contribution of animal models to drug discovery, it remains difficult to conduct mechanistic studies on the barrier function and interactions with drugs at molecular and cellular levels. One innovative approach to addressing this challenge is to develop an in vitro model that mimics the essential physiological structure and function of the human BBB and that allows quantitative analysis of drug transport across the barrier in a controlled manner. The main focus of this thesis is on development of a microengineered human BBB model which reconstitutes the key structure and function of the human BBB and enables 3D capturing of nanoparticle distribution at tissue and cellular levels to demonstrate the mechanisms of cellular uptakes and BBB penetration. This BBB model may present a complementary in vitro model to animal models for prescreening drug candidates for the treatment of CNS diseases.
-
ItemAutomated real-time machine learning for IOT for manufacturing a cloud architecture and API(Georgia Institute of Technology, 2019-11-12) Parto Dezfouli, MahmoudDue to the recent movements in Industry 4.0 and Internet of Things (IoT), accessing or generating data in the Smart Manufacturing (SM) domain has become more attainable; communication protocols such as MTConnect and OPC-UA provide access to a majority of raw data generated from machine tools while retrofit sensor packs facilitate high- frequency data acquisitions from legacy and modern equipment. These technologies have led to the generation of quantities of raw data, known as Big Data (BD), that are complex to be analyzed. Current IoT architectures and frameworks propose Cloud Computing (CC) and Centralized Training (CT) as the addressing solutions for BD and collaborative Machine Learning (ML) models. These solutions, however, have limitations such as Internet dependency and requiring expensive and high-performance cloud resources. As more data are generated, a higher performance framework is required for cloud computing of larger datasets that are either historical in nature or generated from an ever-increasing ubiquitous sensors and sensor arrays that are deployed in modern manufacturing operations. Studying IoT architectures and stream analytics is essential for creation of successful IoT platforms. In this regard, this study proposes a novel, high-performance, and data- driven IoT architecture that considers automated and scalable machine learning techniques with the focus of process control and deeper understanding of manufacturing process and systems performance in the Cyber-Physical Systems (CPS) domain. In this dissertation, first, a novel generalized three-layer IoT architecture utilizing Edge Computing (EC), Fog Computing (FC), CC, and Federated Learning (FL) is presented, where data are preprocessed in the Edge layer, ML models are incrementally trained in the Fog layer and the resulting elements of training are aggregated in the centralized cloud models. Second, two novel stream analytics engines of Outlier Detection and Bayesian Classification, capable of real-time (RT) training and prediction are proposed and analyzed for this architecture. Results show that the training latency for both the Outlier and the Bayesian engines as well as their FL algorithms remained constant as the number of data points increased. On a 1000 data point dataset, the training performances for an upcoming data point for the Outlier and Bayesian engines were on average 136 and 48 times faster, respectively, than retraining the models with all of the data points. These results suggest that the methods discussed in the proposed architecture can lead to the development of higher performance and more scalable IoT frameworks that require lower storage and computing power.
-
ItemFinite element modeling of optic nerve head biomechanics in a rat model of glaucoma(Georgia Institute of Technology, 2019-11-12) Schwaner, Stephen AndrewGlaucoma is the leading cause of irreversible blindness and is characterized by the dysfunction of retinal ganglion cells (RGC), the cells that send vision information from the retina to the brain. All current therapies focus on lowering intraocular pressure (IOP), a causative risk factor in the disease. However, they are not always effective. Although it is well-accepted that elevated IOP-induced biomechanical insult to the optic nerve head (ONH), the region in the posterior eye where RGC axons exit, is key to glaucoma pathophysiology, the mechanisms by which biomechanical insult leads to RGC death are unknown. Rat glaucoma models present an opportunity for understanding glaucoma biomechanics and are widely used in the field. However, rat ONH biomechanics have not been characterized and rat ONH anatomy differs substantially from the human. Therefore, the purpose of this thesis was to provide the first characterization of rat ONH biomechanics to the glaucoma field. To this end, we completed three specific aims. First, we used inverse modeling combined with whole-eye inflation testing to extract material properties from the rat sclera. Second, we conducted a sensitivity study to investigate the effects of anatomical and material property variation on rat ONH strains using a parameterized finite element model of the rat ONH. Lastly, we developed a methodology for building rat ONH FE models with individual-specific geometry and simulated the effects of elevated IOP. Key results include the finding that the patterns of strain in the rat ONH are less symmetric than those in the human, and the highest strains occur in the inferior nerve. In all three aims, the results emphasized the importance of collagen fiber organization on optic nerve strains. Lastly, the patterns and magnitude of optic nerve strain in the parameterized model showed good concordance with those observed in the individual-specific models, suggesting that the higher throughput parameterized models may be able to replace individual-specific models of the rat ONH moving forward. The results from this work can serve to inform future modeling studies on rat ONH biomechanics and provide context for interpreting rat glaucoma studies with the goal of learning more about the link between biomechanical insult and RGC pathophysiology in glaucoma.
-
ItemTransporting fabric: Decentralized multi-agent control through a distributed active environment(Georgia Institute of Technology, 2019-11-12) Motter, KyleThe garment manufacturing industry has largely not benefited from the rapid advances in robotics and automation due to the inherent difficulty in handling flexible materials. At present the vast majority of sewing operations and material handling is still performed by humans in low-wage conditions. However, the industry is undergoing a paradigm shift toward custom and on demand manufacturing, increasing the need for automated handling of cut fabric. This thesis presents a comprehensive system based on novel distributed actuators, called budgers, for fabric manipulation and control. Using these distributed actuators as a foundation, this thesis explores a system architecture to provide practical, factory-ready local fabric control and a scalable solution for routing material through a large-scale implementation. This is presented within the context of treating the fabric as an “unactuated robot” traversing through the “actuated environment” of a budger array. Examined holistically as applied research, the thesis focuses on the actuation, feedback, and control, necessary for robust fabric manipulation. The budger is physically redesigned for significant performance, manufacturability, and serviceability gains. A custom vision feedback algorithm is presented for real-time stable feedback on the state of the fabric, including position and wrinkle information even in the case of deformation or occlusion. And a system architecture for the unactuated robot provides a scalable solution to handling large numbers of fabric across a decentralized network of budger groups.
-
ItemApproaches for the Development of Early-stage Engineering Design Skills(Georgia Institute of Technology, 2019-11-12) Hilton, Ethan C.Early-stage engineering design skills are crucial for the generation and development of solutions to the problems faced by engineers today. Avenues for engineers to develop these critical skills need to be identified and these skills need to be infused into the already packed engineering curriculum. The following work is presented in two studies which aim to develop these key skills. The first study investigates the impacts made on mechanical engineering students who are taught industrial design-based freehand drawing techniques during a six-week, introduction to engineering graphics course. Freehand sketching is an essential skill for communication and visualization in engineering design. The study compares two pedagogies for teaching engineers to sketch: Traditional and Perspective. The Traditional pedagogy contains concepts that have been commonly found in engineering graphics courses. The Perspective version of the course is based on a pedagogy from an industrial design course and contains concepts generally regarded as more advanced sketching skills. Both sketching approaches are found to improve spatial visualization abilities, but there were no significant differences between the groups. The Perspective method was found to be significantly more likely to improve student sketching ability. Thus, the Perspective method was found to be as effective as the Traditional approach for developing spatial visualization skills while developing additional free-hand sketching skills. The second study investigates the impacts of gaining additional prototyping experience through involvement in a makerspace through a longitudinal study of engineering students at three universities. University makerspaces have been touted as a possible avenue for improving student learning, engagement, retention, and creativity. However, there have been very few studies that use empirical data to evaluate how these spaces are impacting the people using them. This study of three university makerspaces measures engineering design self-efficacy and how it is correlated with involvement in the makerspaces, along with student demographics. Students at all three universities are surveyed to determine their involvement in their university’s makerspace and how they perceive their own abilities in engineering design. The findings presented in this paper show a positive correlation between engineering design self-efficacy and involvement in academic makerspaces. Furthermore, correlations are also seen between certain demographic factors and the percentage of students who choose to use the academic makerspace available to them. These findings provide crucial empirical evidence to the community on the self-efficacy of students who use makerspaces and provide support for universities to continue making these spaces available to their students. Together, these studies provide two avenues through which engineers can develop key early-stage engineering design skills such as free-hand sketching and early-stage prototyping. This provides engineering educators with additional tools and resources for how students can be better developed as engineering designers while maintaining the rest of the curriculum.
-
ItemTHERMAL ENERGY TRANSPORT ACROSS ULTRAWIDE AND WIDE BANDGAP SEMICONDUCTOR INTERFACES(Georgia Institute of Technology, 2019-11-12) Cheng, ZheThe development of ultrawide and wide bandgap semiconductors enables a variety of applications in power and RF electronics, including energy infrastructure, wireless communication, self-driving cars, and radar systems for defense. With the increasing power and frequency of these applications, Joule-heating induced hot-spots in the device channel degrade the device performance and reliability. Thermal management of these devices plays a very important role in achieving stable device operation and long lifetime, and correspondingly improving energy efficiency and reducing cost. The basic component, GaN HEMTs, is usually integrated with high thermal conductivity substrates such as SiC and CVD diamond to extract the generated heat. For instance, GaN is grown on SiC with an AlN transition layer. CVD diamond is grown on GaN with an interfacial dielectric layer. The AlN layer and the low quality of the GaN layer near the interface induces additional thermal resistance. The nanocrystalline nature of the CVD diamond near the GaN-diamond interface results in significantly reduced thermal conductivity and large thermal stress due to the high growth temperature and large thermal expansion coefficient mismatch. Additionally, great attention has been focused on β-Ga2O3 recently due to the potential of affordable large-area wafers for homo-epitaxial growth, large breakdown voltages, and its ultrawide bandgap. However, its thermal conductivity is more than one order of magnitude lower than the other wide bandgap semiconductors. A disproportionally small amount of work has been done to address the thermal issues compared to analogous demonstrations of related devices. The understanding of heat transport mechanisms in nanostructures and interfaces, solution to address the thermal management challenges are in demand. The grand challenge of thermal management of power and RF electronics lies in placing the hot-spot area of GaN/AlGaN and Ga2O3 devices close to heat sinks or heat spreaders with small thermal resistance and low stress. Thermal boundary resistance accounts for a large or even dominant part of the total thermal resistance in these devices. This thesis studied the TBC of five technologically important interfaces: GaN-SiC, GaN-diamond, diamond-Si, (Al0.1Ga0.9)2O3-Ga2O3, Ga2O3-diamond. (1) Instead of including a defective AlN transition layer between GaN and SiC in direct growth method, a room-temperature surface-activated bonding technique is used to integrate GaN with SiC which brings high-quality GaN directly to the GaN-SiC interface. The measured GaN thermal conductivity is higher than the MBE-grown GaN on SiC substrates. Moreover, a very high GaN-SiC TBC is observed for the bonded GaN-SiC interface, especially for the annealed interface whose TBC (~230 MWm-2K-1) is close to the highest reported value of GaN-SiC interfaces in the literature. (2) Unlike the growth of CVD diamond on GaN which has a nucleation layer with low thermal conductivity, GaN is heterogeneously integrated with single crystalline diamond substrates with two modified room-temperature surface-activated bonding techniques for thermal management of GaN-on-diamond applications. The measured TBC of the bonded GaN-diamond interfaces is among the highest values reported in the literatures and is affected by the thickness of the interfacial bonding layer. Device modeling shows a relatively large GaN-diamond TBC value (>50 MW/m2-K) achieved in this work could enable device designers to take full advantage of the high thermal conductivity of single crystalline diamond. (3) To improve the low TBC of diamond related interfaces because of the large phonon density of states mismatch of diamond and other semiconductors, the TBC at semiconductor-dielectric interfaces isincreased by nanoscale graphoepitaxy. By growing CVD diamond on nanopatterned silicon wafers, a general strategy is provided to significantly reduce the thermal resistance of both a diamond layer and diamond-substrate interface simultaneously. The diamond-silicon TBC could increase by 65% comparing with that of a flat diamond-silicon interface. (4) To understand the phonon transport mechanisms across β-(Al0.1Ga0.9)2O3-Ga2O3 interfaces, temperature-dependent measurement on thermal conductivity of β-(Al0.1Ga0.9)2O3/Ga2O3 superlattices is reported from 80 K to 480 K. Significantly reduced thermal conductivity is observed (5.7 times reduction) at room temperature comparing with bulk Ga2O3, which highlights the importance of thermal management of related devices. The estimated minimum TBC of β-(Al0.1Ga0.9)2O3/Ga2O3 interfaces is found to be larger than the Ga2O3 maximum TBC, which shows that some phonons could transmit through several interfaces before scattering with other phonons or structural imperfections, as possible evidence of phonon coherence. (5) To develop cooling strategies of Ga2O3-related devices, Ga2O3 is integrated with single crystal diamond with exfoliation-transferring and ALD-growth. The Van der Waals Ga2O3-diamond TBC was measured to be 17 -1.7/+2.0 MW/m2-K while the TBC calculated with a Landauer approach and DMM is 312 MW/m2-K, which sheds light on the possible TBC which can be achieved. The measured TBC of the grown ultra-clean interface is 179 MW/m2-K, about 10 times higher than TBC of a Van der Waals bonded Ga2O3-diamond interface, suggesting that covalent bonding facilitates interfacial heat transport better than Van der Waals interfacial bonding. Integration of Ga2O3 and single crystal diamond could be a solution to cool Ga2O3-related devices.
-
ItemDistillation assisted purification, transport and delivery of liquefied natural gas(Georgia Institute of Technology, 2019-11-12) Manek, VeeraThe removal of dissolved CO2 from natural gas is essential for the safe and reliable operation of liquefied natural gas (LNG) systems. The purification of natural gas (NG) from CO2 down to a concentration of 50 ppm by multi-stage distillation is theoretically investigated. A three-column distillation system is proposed that can purify NG to lower than 50 ppm concentration of CO2, while avoiding CO2 freezeout. The columns include a 30-stage Demethanizer, in which high purity methane is obtained in the distillate by separating the impurities from natural gas including CO2; a 50-stage extractive column where the azeotrope between CO2 and ethane is broken; and a 50-stage solvent recovery column that recovers a mixture of heavy hydrocarbons suitable for recycling as a solvent back into the extractive column. The proposed system avoids CO2 freezeout by utilizing a multi-component feed of some heavier hydrocarbons added to natural gas; propane, butane and pentane additives are injected into stage 20 of the Demethanizer column alongside the raw feed. Furthermore, arrangements are made to break the CO2-ethane azeotrope, which may occur in the bottoms stream of the Demethanizer by administering a solvent stream in the extractive column. The proposed system can operate in a closed-loop arrangement where the bottoms stream that leaves the recovery column can be recycled and injected into the extractive column for azeotrope prevention. Hydrodynamic and heat transfer characteristics of a double helically coiled tube confined in a cylindrical shell is experimentally studied using an instrumented test loop that represents a prototypical LNG fuel delivery system for natural gas-burning IC engines. The test loop comprises of a heat exchanger consisting of a double-helically coiled tube that carries liquid nitrogen (liquefied natural gas (LNG) in the real system), placed in a shell-confined secondary side through which a secondary coolant (a mixture of propylene glycol and water in the experiments, and engine oil in the prototype) flows. Experiments addressing liquid (water) and gas (nitrogen) single-phase flows, as well as two-phase flows (air-water), are performed. CFD simulations are carried out, and empirical correlations are developed for the frictional pressure losses and two-phase pressure multiplier for the double helically coiled heat exchanger.
-
ItemStability of self balancing transporters(Georgia Institute of Technology, 2019-11-12) Castro Castro, Arnoldo GerardoA class of personal transporters based on Two-wheeled inverted-pendulum machines has emerged as alternative transportation system for urban and indoor environments. However, these machines are inherently unstable. Typical use conditions can lead to very unstable and dangerous conditions. Furthermore, the control system does not attempt to stabilize the system laterally, which creates a fast-acting hazard that is extremely challenging for the user to anticipate and mitigate. In this work, a vehicle and user model will be developed to investigate the potential operating parameters that can cause failure during normal use. This model will aim to provide a more thorough description of the wheel-ground interaction and the dynamics of a human rider than previous works. The models will also serve as a test platform for evaluating traction-control methods that have not typically been used in these devices and that can potentially ameliorate the dangers associated with wheel slipping and lateral instability.