12023 research outputs found
Sort by
Sea Ice Discrete Element Modeling: Melt and Fracture of Floes and Sheets
Get PDFOver the last 40 years, the Arctic Ocean has experienced a significant reduction in surface area and thickness of sea ice for its minimum summer and year-round values. Sea ice, existing both as continuous ice sheets and distinct broken floes or blocks, is disappearing earlier and faster over time. These changes are largely occurring within marginal ice zones, where ice is most vulnerable to thermal forcings from the sun, oceans, and atmosphere and wind and ocean currents. Given that sea ice plays a vital role in regulating climate by delaying global energy exchanges, its loss is a vital factor in increasing global temperatures and the frequency of extreme weather events. Understanding and projecting seasonal variations in sea ice is imperative to improve climate predictions. However, many of the processes in sea ice are not fully described by most existing models, due to the limitations of continuum sea ice approaches. As a result the use of discontinuum techniques on sea ice is a very active field. In this work, we combine discrete element methods with satellite image analysis to study changes in sea ice concentration and floe size distribution during the summer melt transition for ensembles of distinct floes decaying into open waters and continuous fast sea ice sheets breaking into multiple floes. For the pure floe-like behavior, we use the `Level Set Discrete Element Method for Sea Ice' or LS-ICE method. This model can resolve individual sea ice floes with realistic shapes, and represent their physical interactions by leveraging level-set functions to detect contacts. LS-ICE can also be coupled to atmospheric and oceanic heat and momentum forcings and simulate associated melt and breakage processes. With it, we are able to reproduce sea ice concentration decline for the summers of 2018 and 2020 at Baffin Bay. Using LS-ICE we also unveil the sensitivity of sea ice loss and floe size distribution to different intensities of fracturing and ocean/solar melt and how sea ice floe size determines which is more dominant. For monolithic landfast sea ice sheets, we use a bonded particle method within the level set discrete element model called LS-DEM-BPM. We explore the relationship between landfast sea ice breakage and area decline, ocean currents and floe size distribution for a region in Fram Strait in 2023. We also replicate its fracture characteristics, using idealized pulses and arbitrary eddying ocean currents, and unveil particular combinations of wavelengths and wave speeds that facilitate breakage. Our results give new insight on sea ice melt and breakage interactions and provide a numerical framework for simulating the complete transition of sea ice from intact sheets to open oceans
Using Functional Genomics to Characterize Biogenesis and Quality Control Pathways in the Mammalian ER
No full textCells are tasked with ensuring the proper synthesis, localization, and insertion of membrane proteins at the mammalian endoplasmic reticulum (ER). Recent advances have shown that the insertion of transmembrane domains into the ER and the translocation of their associated soluble domains across the ER can be facilitated by members of the Oxa1 superfamily of insertases. The ER membrane protein complex (EMC) contains an Oxa1 insertase that facilitates the co-translational insertion of the first TMD of Nexo proteins, which have their N-terminal soluble domains localized in the ER lumen or extracellular space. Additionally, recent work has described the multipass translocon, a supercomplex at the Sec61 translocation channel that facilitates insertion of multipass membrane proteins. During my PhD and in collaboration with other scientists in the Voorhees lab, I elucidated how the EMC cooperates with the multipass translocon to facilitate biogenesis of multipass membrane proteins. We took a systematic approach, applying a combination of functional genomics, biochemistry, structural biology, and mechanistic cell biology to understand how the biophysical properties of the TMDs and the intervening soluble domains of multipass membrane proteins influence their path into the ER bilayer. We show that the EMC is epistatic with members of the multipass translocon, including the BOS, GEL, and PAT complexes. We structurally characterize the EMC•BOS holocomplex, showing that these complexes directly interact, and that this interaction is mutually exclusive to the interaction of BOS and Sec61. Further, we demonstrate that Nexo proteins that contain a net positive charge in their N-terminal soluble domain are difficult for the EMC to insert and thus also rely on Sec61 or TMCO1, the Oxa1 insertase of the GEL complex, for insertion. We overturn the prevailing model for multipass membrane protein insertion and show that how this diverse class of membrane proteins utilizes the suite of ER biogenesis machinery depends on their distinct biophysical properties. In addition to biogenesis, during my PhD I also studied how the cell surveils multi-subunit complex assembly. I focused on a model ER-resident and obligate complex. These subunits are unstable and degraded in the absence of their binding partner, but how they are recognized for degradation is unknown. I used unbiased functional genomics approaches to identify the quality control components that regulate orphan subunit degradation in the cell. Further, I used proteomics to identify endogenous substrates of this particular ERAD pathway. This work expands upon our understanding of how multi-subunit complexes are regulated by machinery in the cell
Scattering and Gravitational Effective Field Theory
No full textAdvances in the methodologies developed in quantum field theory and in the scattering amplitudes program have led to their application to questions pertaining to the classical physics of gravitationally interacting binary systems. The perturbative and relativistic nature of the quantum field theoretic setup is perfectly suited for obtaining results in an expansion in the gravitational constant, also known as the post-Minkowskian (PM) expansion. However, there are several practical scenarios where the gravitational waves produced by the inspiral or interaction of two massive bodies arise from dynamics in the strong field regime and the PM expansion breaks down. Extreme mass ratio inspirals, where a lighter body interacts with a much heavier black hole, are examples of such systems.
In contrast, classical solutions, such as the Schwarzschild metric, and the geodesic trajectories of test bodies traversing in these nontrivial backgrounds encode information to all orders in the gravitational constant. In fact, these solutions can be viewed as the summation of certain infinite sets of Feynman diagrams from the perspective of point particle effective field theory (EFT). Alternatively, metrics and related geodesic trajectories can be seen as performing enormous simplifications of the tensor structures arising in these equivalent sets of Feynman integrals. We describe how the all order in PM information present in classical solutions can be utilized to simplify PM calculations in point particle EFT and set up a systematic framework for studying the classical dynamics of binary systems as an expansion in their mass ratio.
We also delve into questions about the origin and scope of validity of color-kinematics duality and the double copy relation, which can be used to generate amplitudes of one theory from another. For example, graviton amplitudes can be obtained from gluon amplitudes. Unveiling the underlying structure that gives rise to these relations would not only deepen our understanding of the properties of these theories but could also serve in streamlining their application to computations of practical interest such as those showing up in the study of the gravitational two-body problem using field theory techniques. Specifically, we analyze a toy system in two dimensions where we find a Lagrangian-level manifestation of the duality in a classical equivalent of the nonlinear sigma model. We unpack the implications of an off-shell formulation of the color-kinematics duality and double copy in order to understand the possible wider implications for these relations in other theories.</p
Mountains of Self: Dream Revelations Transcending Frame Narratives
No full text[Introduction] Dreams have long been a source of curiosity and awe to humankind, inspiring centuries of study, interpretation, and discussion. The enigma of dreams—where they come from, what they mean, and why they occur—is shared by nearly all people, so it’s no surprise that the concept of dreaming has also been the subject of countless stories. Stories told through lens of both dream and reality often use framing as a method of separating the narrative of the dream from the narrative of “reality” in the story. In such stories, there are usually three frames: the frame of the reader and the world the reader is living in, the frame of the story, and the frame of the dream within the story. Such stories interrogate the connection between these three frames; almost invariably, the connection is related to dreamer themselves. Dreams manifest truths about the dreamer, often revealing a picture of a self that the dreamer may not be able to see or face in waking life. Elements of a dream can be manifestations of these truths; these elements then transcend the frame of the dream, reaching from the dream into the story’s reality and, at times, even into the reality of the reader themselves
Vibration Damping of Coiled Structures Through Frictional Slip
Get PDFVibration management is important for the survivability of structures. The response of a structure under vibration is dependent upon interaction between the excitation environment and the properties of the structure. If the input excitation cannot be adjusted, then the structure must be engineered to survive. One approach to engineering structures to reduce vibration response is through damping, which is achieved by adding damping devices or materials to covert kinetic energy into heat, where removing energy from the system reduces the amplitude of response. There are a variety of existing vibration damping concepts and techniques, however, conventional methods of these approaches are subject to limitations such as compromising stiffness for increased damping and performance that is excitation profile dependent.
This research proposes a novel, passive vibration damping concept which is motivated by recent deployable structures for space that use coiling as a packaging architecture. The proposed concept, referred to as "wound roll damping", is a friction-based damping scheme for coiled structures, where the structure is wound around a mandrel with tension that allows interlayer slip during vibration. The friction between slipping layers provides an energy dissipation mechanism, which reduces the overall level of response. The concept was developed with the challenges of mitigating spacecraft launch vibration and the limitations of conventional damping techniques in mind.
Understanding of the working principle and performance of this damping concept is achieved using a combination of experiments, analysis, and FEA. A method for determining the locations of slip within a wound roll under vibration is presented. This consists of modeling the interlayer friction forces, using analytical expressions for the stress fields that arise during tension winding of wound rolls, and comparing these values against loading estimates obtained from analysis and FEA. The locations of slip for wound rolls supported by a cantilevered mandrel with bending vibration modes are towards the root of the wound roll structure, near the inner layers.
Experimental studies that demonstrate the performance and properties of this damping concept are presented in this work. A wound roll test sample is subjected to a range of excitation profiles including: sine sweep, sine dwell, random, and shock with varying levels of sample winding tension and excitation amplitude. Using these experiments, this concept is demonstrated to not be subject to the limitations of conventional damping schemes. This scheme is observed to be capable of significantly increasing the overall stiffness while providing elevated damping levels, with a performance that is tunable with winding tension, independent of excitation profile, and scales with excitation amplitude. The locations of slip are observed to be consistent with predictions from FEA and analysis.
Two approaches to simulate and model the wound roll damper are developed to both better understand the physical mechanism of this concept and provide analysis tools. The first method is an FEA model, consisting of the base vibration of concentric shells and solids that have frictional contact interactions. The second method is a 2-DoF reduced order model that simulates the frictional contact between two mass-spring-damper systems. Both methods are demonstrated to have good correlation with experimental measurements.
A majority of this work demonstrates the performance of this concept, using both experiments and simulation at lab scales. This work also presents simulation studies that demonstrate the viability of this concept at realistic scales. Using simulations scaled to recent coilable space structures, both implemented and proposed, the wound roll damping concept is demonstrated to provide significant stiffness and damping.</p
Investigation of Transport Phenomena in Semiconductors and Semiconductor Devices: Drain Noise, Two-Phonon Scattering, and Phonon Drag
Get PDFThe dynamics of charge carriers in semiconductors set the foundation for semiconductor device performance. Devices crucial for fields like radio astronomy rely on transistor amplifiers where hot electron dynamics impact noise significantly. The overarching goal of this work is to contribute towards the development of better transistor amplifiers by investigating electron transport in existing devices and emerging materials.
The physical mechanisms governing noise in a class of semiconductor devices called high electron mobility transistors (HEMTs) are not completely understood. HEMTs are transistors that use a junction between two materials of different band gaps as the channel. HEMTs are used as amplifiers by translating a small signal applied at the gate terminal to a large current at the drain terminal or output. The noise added at the input is well-characterized by the device physical temperature, while the origin of the noise added at the output is still up for debate. We attempt to fill this knowledge gap by proposing a theory of noise occurring at the drain terminal of these devices as a type of partition noise arising from two possible electron paths. This theory emphasizes the critical role of the conduction band offset between epitaxial layers of the device: a larger offset maximizes the channel sheet density and minimizes electron transfer between layers, potentially improving noise performance. The theory accounts for the magnitude and dependencies of the drain temperature and suggests strategies to realize devices with lower noise.
We then investigate phonon-limited charge transport in the semiconductor boron arsenide. Boron arsenide has drawn significant interest due to reports of simultaneous high thermal conductivity and ambipolar charge mobility, desirable properties for integration in electronic devices. The theoretical prediction of high electron and hole mobility assumed the dominance of charge carrier scattering by one phonon. We consider the effects of two-phonon electron and hole scattering processes in boron arsenide, and find that inclusion of these higher-order processes reduces the computed room-temperature electron and hole mobility significantly from the one-phonon value. Despite its potential, our predictions of electron and hole mobility contradict recent experimental reports based on photoexcited charge carrier diffusion. Several factors may explain this discrepancy, including another type of two-phonon scattering not considered in this work, superdiffusion of hot carriers, induced carrier concentration, or a combination of all or some of the above elements.
At high carrier concentrations, the phonon system may interact with the electron system on the timescale of the phonon-phonon interaction. When this happens, the nonequilibrium state of phonons becomes important for electron transport, and vice versa as these systems interact in a coupled manner. This coupled interaction could lead to an inflated value of the experimentally reported mobility. We quantify this effect, known as phonon drag, with a coupled electron-phonon Boltzmann transport equation framework and demonstrate that the electron mobility is indeed enhanced significantly at the relevant carrier densities.</p
Charge Transport Phenomena in Cryogenic SiGe Heterojunction Bipolar Transistors
Get PDFSilicon-germanium heterojunction bipolar transistors (HBTs) are widely used for high-speed communications and radar systems owing to the their low-cost and competitive performance relative to III-V compound semiconductor devices. Due to the higher cost and lower yield of III-V high electron mobility transistors (HEMTs) based on InGaAs quantum wells, SiGe HBTs operating at cryogenic temperatures are of significant interest for radio astronomy and quantum computing. However, their microwave noise performance has long been observed to be poorer than those of HEMTs. As a result, the physical mechanisms governing the cryogenic DC, microwave and noise performance of SiGe HBTS have been a topic of investigation for many years. Improved understanding of these mechanisms may ultimately allow for the realization of HBTs with noise performance rivaling those of HEMTs yet with lower cost, improved compatibility and integration with CMOS processes, and high yield.
This thesis uses theoretical and experimental methods to examine cryogenic charge transport phenomena in SiGe HBTs which affect the microwave noise performance. A particular focus is on the anomalous electrical characteristics at cryogenic temperatures, in which pronounced deviations from the ideal drift-diffusion theory are observed. Various explanations for the observed anomalous cryogenic I-V behavior have been postulated, such as quasi-ballistic transport and electron tunneling, among others. Despite a number of works on this topic over the past three decades, none of the explanations has been unambiguously confirmed or excluded.
The first contribution from this thesis is a study of the quasiballistic transport hypothesis using an exact, semi-analytic solution of the Boltzmann equation. Several prior studies have claimed quasiballistic electron transport across the base as the origin of cryogenic non-ideal current-voltage characteristics. Specifically, the observation of temperature independent DC performance below K has been attributed partly to quasiballistic transport resulting in a presumed increase in electron temperature, but this hypothesis has been examined only using empirical models which leave ambiguity. We overcome this limitation by adapting an exact, semi-analytic solution to the Boltzmann equation based on an asymptotic expansion approach to describe electron transport across the base region of an HBT. With this exact solution, we computed macroscopic electrical properties such as collector current and transconductance which could be directly compared with experiments. We find that the computed transport characteristics are inconsistent with experiment, with the calculated transconductance following the ideal drift-diffusion inverse temperature dependence. This finding implies that quasiballistic electron transport is unlikely to be the origin of cryogenic non-ideal I-V characteristics.
Next, we study a previously unexplored explanation, the presence of lateral spatial inhomogeneities in the base-emitter junction potential height, as the origin for the observed non-ideal cryogenic current-voltage anomalies in SiGe HBTs. While this phenomenon has been established as the origin for similar cryogenic I-V anomalies observed in Schottky diodes, this possibility has not yet been considered for SiGe HBTs. We experimentally investigate this hypothesis by characterizing the base-emitter built-in potential and its temperature dependence using both capacitance-voltage and current-voltage characteristics. We observe a marked discrepancy in the built-in potential obtained using these two methods at cryogenic temperatures, a signature consistent with the presence of lateral inhomogeneities in the junction potential. We hypothesize that these inhomogeneities arise from clustering of Ge as a result of aggressive doping of modern devices, and propose future directions that allow direct probing of these inhomogeneities.
Finally, we explore the potential improvements in the minimum achievable noise temperature of HBT amplifiers by considering the effects of shot-noise correlation. We first model the expected reduction in cryogenic noise temperature of a state-of-the-art transistor as a result of shot-noise correlation. We then quantify the accuracy of the present noise measurement techniques that allow us to exploit the benefits of shot-noise correlation, and propose modifications to the noise measurement setup that will permit an unambiguous experimental determination of the magnitude of the effect.</p
The Impact of Energy Availability and Substrate Complexity on Anaerobic Microbial Communities in Marine Sediment
Get PDFThis thesis probes the interplay of organic matter complexity (Chapters 1 and 2) and local redox gradients (Chapter 3) with the community structure and function of the anaerobic marine sediment microbiome. Deep marine sediments, despite being generally organic-poor, harbor a vast diversity of microorganisms that are critical to the global nutrient cycle. Transient nutrient inputs such as whale falls result in hotspots of microbial community activity in an environment that normally processes heavily degraded organic material from the upper ocean. These organic loading events result in transitions down redox gradients and dynamic shifts in the local energy availability of the microbial communities. Through in situ seafloor and laboratory microcosm experiments, we provide insights into the impact of energy availability and carbon complexity on maintaining hierarchical and complex community interactions, community activity, and systematic and functional diversity
Altering Framework Topology and Heteroatom Distributions of Molecular Sieves by Designed Organic Structure-Directing Agents
No full textThe growing demand for chemical production combined with the urgent need to mitigate the accelerated climate and environmental changes motivates efforts to create highly efficient and selective catalysts and adsorbents. Zeolites and molecular sieves are a key class of materials for addressing these needs because of their high activity and selectivity with catalytic reactions. Additionally, they can show superior adsorption properties because of their structure and surface polarity that can also give shape selectivity with molecules smaller than ca. 1 nanometer. Further advancements in molecular sieve properties will rely on advancements in preparation methods. To this end, the research results presented here explore synthetic approaches for controlling the framework topology and the heteroatom incorporation within silicate-based molecular sieves by means of the strategic design of their organic structure-directing agents (OSDAs).
Part I presents the synthesis of STW-type germanosilicate molecular sieves with high-silica framework compositions and the enrichment of chirality. A chiral OSDA is computationally designed based on the predicted stabilization energy toward the pure-silica STW framework. An improved synthesis route for both enantiomers of the OSDA is developed. The enantiopure OSDA is capable of crystallizing a high-silica STW-type germanosilicate molecular sieve that shows distinct framework compositions from previously reported germanium-rich STW. The enantiomeric enrichment of powdered samples without occluded enantiopure OSDAs is characterized by the dynamical refinement of microcrystal electron diffraction data. The high-silica, enantiomerically enriched STW exhibits the framework stability upon thermal treatment and the enantioselective adsorption of 2-butanol. The results in Part I demonstrate the design strategy of OSDAs for crystallizing stable, enantio-enriched molecular sieves for enantioselective chemical separations and catalysis.
In Part II, the distribution of heteroatoms incorporated within borosilicate molecular sieves is studied with regard to its control by cationic OSDAs. To aid in the characterization of the heteroatom sites within borosilicate molecular sieves, the relationship between the 11B NMR chemical shift and the local geometry of boron within tetrahedrally coordinated silicate frameworks is first investigated. From crystalline borosilicate minerals with highly ordered, tetrahedrally coordinated boron atoms, it is revealed that the chemical shifts from 11B NMR linearly correlate with the local geometric parameters. Further studies on the borosilicate molecular sieves that possess more open space and wider angles suggest that the correlation between the average bond angles and 11B NMR chemical shifts can be employed for the entire class of three-dimensional, crystalline borosilicates. Two structurally similar quaternary ammonium OSDAs with different locations of positive charge are designed and synthesized. MWW-type borosilicate molecular sieves are crystallized by both OSDAs, and the quaternary ammonium moieties in the two OSDAs are found to interact with boron species with significantly different 11B NMR chemical shifts. Using the correlation developed here, the characterization results demonstrate that the heteroatom siting within the molecular sieve framework can be selectively altered by tailoring the OSDA structure in terms of the position of positive charge.</p
Learning in the Quantum Universe
Get PDFIn this thesis, I will present our progress in building a rigorous theory to understand how scientists, machines, and future quantum computers could learn models of our quantum universe. The thesis begins with an experimentally feasible procedure for converting a quantum many-body system into a succinct classical description of the system, its classical shadow. Classical shadows can be applied to efficiently predict many properties of interest, including expectation values of local observables and few-body correlation functions. I will then build on the classical shadow formalism to answer two fundamental questions at the intersection of machine learning and quantum physics: Can classical machines learn to solve challenging problems in quantum physics? And can quantum machines learn exponentially faster and predict more accurately than classical machines? The thesis answers both questions positively through mathematical analysis and experimental demonstrations