12023 research outputs found
Sort by
The Influence of Négritude and Black Postcolonial Thought on Derek Walcott’s The Odyssey: A Play
No full text[Introduction] Derek Walcott’s The Odyssey: A Play is an homage to both Homer’s Odyssey and the anti-colonial Négritude literary movement, which aimed to spread African cultural values and the centering of Black identity to the broader African diaspora. Walcott modifies the original Odyssey to include elements of his own St. Lucian heritage through the strategic addition of Afro-Caribbean language, characters, such as the narrator Billy Blue, spiritual traditions, and more modern anti-colonial ideas developed by the African diaspora. However, Walcott’s attempt to weave together postcolonial modes of thought and Homer’s Odyssey fails to capture the limited depth Homer imbues in his female characters. The shortcomings of the Négritude movement with respect to women, particularly women of color, are seen in the way Walcott writes nonwhite female characters, such as the slave Eurycleia and the enchantress Circe, as either caretakers (“mammies”) or seductresses (“jezebels”), stripping them of their agency and reducing them to stereotypes
The Role of Context-Dependent Metabolic Interactions in Organizing Microbial Communities
Get PDFWe can image the strikingly beautiful compositions of natural microbial communities, but we still lack an understanding of the factors that shape their organization. Understanding the drivers of these structures at the microscale may allow us to better predict and control large-scale community functions in dynamic environments. In this thesis, I developed quantitative image analysis pipelines for uncovering the spatiotemporal growth of aggregate biofilms within a developing oxygen gradient by expanding upon the Agar Block Biofilm Assay (ABBA). I then developed the Agar Disk Biofilm Assay (ADBA) for improved imaging resolution. These tools push the bounders of laboratory experiments to better capture the complexity of natural environments. Next, I built a synthetic microbial community reflecting a metabolic pathway often partitioned between members found in nature: Pseudomonas aeruginosa (PA) strains with a denitrification pathway genetically split at the nitric oxide (NO) node. I characterized the growth of a strict consumer and a strict producer of NO and found that PA metabolizes NO in a manner that supports growth, a previously underappreciated energy conservation strategy. Local oxygen flips this interaction from beneficial to detrimental by increasing toxicity. I found these principles drive context-dependent cellular organization. This work underscores the contributions of partitioned metabolic pathways, redox-active metabolites, and dynamic micro-niches to the organization of microbial communities. Finally, combining my efforts towards method development and an appreciation for how redox-active metabolites drive context-dependent microbial interactions, I show how phenazines promote a previously unrecognized form of slow growth under nutrient limited environments. Taken together, this thesis highlights the importance of understanding dynamic micron-scale microbial interactions and presents several methodological improvements to capture it
Agriculture and Its Role in the Global Carbon Cycle
Get PDFCrops not only feed the world's human population and livestock but also impact the global carbon cycle. The intensification of agriculture has allowed much greater crop yields by hybridization, irrigation, and fertilization in the five most recent decades. However, the increased frequency and severity of extreme weathers (e.g., heat wave, drought, flood) caused by global warming have led to large yield and economic losses. Thus, the monitoring of crop growth in a changing climate is of paramount importance to improve food security and alleviate poverty. It is via photosynthesis that crops use the energy of sunlight to reduce carbon dioxide (CO₂) into carbohydrates. An accurate quantification of plant photosynthesis is a key step towards estimating crop yield and understanding the carbon exchange between the biosphere and atmosphere. Satellite remote sensing has emerged as one promising solution for measuring photosynthesis from regional to global scales. In the thesis, first, we show the potential of solar-induced chlorophyll (SIF) signals emitted by the chlorophyll a of plants to track photosynthesis. Compared to traditional reflectance-based vegetation indices (VIs), SIF can better capture photosynthetic down-regulation under drought and heat stresses due to its physiological linkages with photosynthetic processes. Second, we demonstrate that SIF can be used to estimate crop yield. At field sites, we find a high correlation between SIF and crop photosynthesis measurements. Scaling up this relationship to the large scale, we show that crop yield estimates using satellite-derived SIF agree well with the United States Department of Agriculture (USDA) reported annual crop yield. Third, we examine how crops respond to climate change and air quality in China. We develop a crop yield prediction model, based on a large volume of historical crop data, as well as climate and pollution records. Our finding demonstrates the co-benefit of the recent air pollution control policy from an agriculture and food perspective. However, such a benefit will be significantly offset or even outweighed by continuing global warming. Fourth, we focus on how different ecosystems, especially intensified agriculture, has reshaped the seasonality of atmospheric CO₂. Our satellite-derived global terrestrial carbon fluxes capture the observed CO₂ seasonal cycle amplitude (SCA) trends at surface sites very well. We further find that CO₂ SCA trends at mid latitude sites around the Midwest United States are mainly impacted by intensified agriculture, whereas high latitude sites are mainly driven by increasingly productive natural ecosystems. The approaches, findings and datasets developed through the thesis will contribute to agro-ecosystems management in the face of climate change and contribute to equitable solutions to climate challenges.</p
Digital Quantum Simulation of Physical Systems on Noisy Intermediate-Scale Quantum Computers
Get PDFCurrent quantum computers are characterized as having the order of 5-100 qubits, with limited connectivity restricting two-qubit operations to nearest neighbors, and with too much noise to achieve fault-tolerance. Such devices, called noisy intermediate-scale quantum (NISQ) devices, have been demonstrated to have sufficient coherent lifetime to perform interesting experiments motivated by quantum information sciences. This motivates the question of whether such devices can be utilized to study physical systems commonly encountered in condensed matter and quantum chemistry.
In this thesis, we address the open problem of identifying approaches to perform quantum simulations of physical systems on NISQ devices. We begin our study by considering the Hamiltonian ground state problem, a task routinely solved in numerical studies of materials and molecules. We provided a new quantum primitive, the quantum imaginary time evolution (QITE), that provides a practical approach to solve the Hamiltonian ground state problem. In addition, the QITE subroutine can be used in a Lanczos scheme to speed up convergence time.
Next, we consider the problem of performing finite temperature simulations and demonstrate how QITE can be used as a subroutine to develop scalable and feasible approaches to perform such calculations on a quantum computer. More specifically, we develop routines to obtain thermal averages by sampling minimally entangled thermal states, and also free energy by evaluating the partition function directly.
In our final study, we consider the study of topological states of matter, which do not fit within the Landau paradigm of local order parameters associated with symmetry breaking, and have been shown to exhibit unusual behavior. We show how a specific class of topological states of matter, the symmetry-protected topological states can be feasibly realized on present NISQ devices and their unusual behavior experimentally validated. Our study provides a benchmark of capabilities of state-of-the-art NISQ devices to study these interesting phases of matter.</p
Strategies for Enabling Stable and Efficient (Photo)Electrochemical Water Splitting
No full textThe electrolysis of water splits H₂O into its constituent parts, generating H₂ fuel and O₂ as a by-product. Although electrolysis has been known since late 1700s and has a consistently expanding industrial capacity, several barriers still exist to its widespread utilization as a clean method of generating hydrogen for industrial uses or as a grid-scale energy storage chemical. Among these, the materials and costs constraints surrounding the use of precious metal catalysts and expenses associated with balance-of-system costs are of primary importance. In this thesis, the first point is addressed by utilizing earth-abundant catalysts for chemical, electrochemical, and photoelectrochemical water splitting reactions. Specifically, MnySb1-yOx catalysts were synthesized for use as both cerium-mediated chemical water oxidation catalysts and as electrochemical water oxidation catalysts, furthering steps towards removing Ir from industrial electrolysis devices. Addition of Sb was shown to stabilize reactive Mn centers in these configurations, offering enhanced stability over pure Mn oxide catalysts. Reduction of electrolyzer balance-of-system costs were addressed in this thesis through the integration of multiple components of a solar-powered electrolysis system into a single, integrated photoelectrochemical water splitting device. Specifically, electrodeposition conditions were shown to affect the spontaneous mesostructuring of Ni-P hydrogen evolution catalysts on silicon photocathodes, leading to enhanced transmission of light to the semiconductor substrate. Furthermore, Y₂SiO₅ protective layers were shown to mitigate the corrosion of Si photocathodes in alkaline environments, an electrochemical environment known to be destructive towards silicon
The Representation of Multimodal Tactile Sensations in the Human Somatosensory System
Get PDFThe sense of touch is critical to executing basic motor tasks and generating a feeling of embodiment. To construct touch percepts, the brain integrates information from tactile mechanoreceptors with inputs from other senses and top-down variables such as attention and task context. In this thesis, we investigate how these factors influence neural activity within the somatosensory system at different stages of tactile processing, using electrophysiological and behavioral data from a human tetraplegic participant implanted with microelectrode arrays. First, we find that neural responses to imagined touches of different types are decodable in the primary somatosensory cortex, ventral premotor cortex, and the supra-marginal gyrus, and these responses remain stable over many months. Following this analysis, the primary somatosensory cortex is explored in greater depth to better characterize early-stage cortical tactile processing. Touches to the arm and finger are examined during a passive task, in a variety of conditions including visually observed physical touches, physical touches without vision, and visual touches without physical contact. Analysis of the two touch locations suggests that touch encoding in primary somatosensory cortex may be less rigid than in the classical topographic view. Additionally, this experiment uncovers a modulatory effect of vision in the primary somatosensory cortex when it is paired with a physical touch, but no effect of vision alone. Finally, we investigate how visual information impacts artificial tactile sensations, which can be elicited using intra-cortical microstimulation to the primary somatosensory cortex. The ability to elicit reliable, naturalistic artificial touch sensations is vital to the implementation of a tactile brain-machine interface, which would benefit patients with spinal cord injury and others with somatosensory impairments. We find that visual information biases the qualitative percept of artificial stimulation towards an interpretation that is visually plausible. The temporal binding window between vision and stimulation is found to be larger when visual information is biologically relevant, suggesting that the brain’s ability to causally relate artificial stimulation to visual cues depends on visual context. Additionally, recordings from the primary somatosensory cortex indicate that visual information relevant to artificial stimulation is represented across contexts, during an active task. The effect of task on the responsiveness of the primary somatosensory cortex to visual information points to a role of attention in mediating early cortical tactile processing. In combination, the findings presented in this thesis provide insight into the basic neuroscience of how tactile experiences are constructed by the brain, suggesting that early tactile processing is influenced by multisensory, contextual factors. These findings also have clinical applications to developing a brain-machine interface capable of providing naturalistic sensations within a complex real world environment
Towards Universal Integrated Laser Sources with Nonlinear Photonics
Get PDFLasers are ubiquitous in modern technology with different applications typically requiring different laser wavelengths. However, a given laser can operate only in a relatively narrow spectral region given by the particular material used to build the laser. This leads to using several lasers when several wavelengths are required. Nonlinear photonic devices pose a solution to this problem by transferring energy from single lasers to vast regions of the electromagnetic spectrum. But, despite more than 60 years of development in nonlinear photonics, most nonlinear devices remain large, expensive, and confined to research laboratories.
In this dissertation, we demonstrate a new generation of integrated nonlinear photonic devices based on the quadratic χ(2) nonlinearity. Using the up-and-coming thin-film lithium niobate platform, we demonstrate ultrafast optical parametric amplifiers, parametric generation of ultrashort mid-infrared pulses, long pulses and frequency combs tunable over an octave bandwidth, and the first χ(2) CW parametric oscillator directly pumped by a single commercial diode laser. These results represent key milestones towards compact and inexpensive universal laser sources.</p
On the Role of Three-Dimensional Genome Organization in Gene Regulation and mRNA Splicing
Get PDFThe nucleus is spatially organized such that DNA, RNA, and protein molecules involved in shared functional and regulatory processes are compartmentalized in three-dimensional (3D) structures. These structures are emerging as a paradigm for gene regulation, a highly complex process that requires the dynamic coordination of hundreds of regulatory factors around precise targets in different cell states. We describe the discovery of hundreds of RNA-DNA hubs throughout the nucleus that are organized around essential nuclear functions such as RNA processing, centromeric heterochromatin organization, and gene regulation. Focusing on RNA processing, specifically co-transcriptional splicing, we find that genome-wide organization of active genes near nuclear speckles drives the efficiency of pre-mRNA splicing in a cell-type specific manner. The results of this thesis illustrate how spatial compartmentalization of biomolecules increases the local concentration of reactants and enzymes such that greater efficiency is achieved in scenarios where rapid responses are required for cell survival
Experimental Study on the Thermodynamic Interactions of Phonons and Magnetism in Fe Systems
Get PDFThe macroscopic thermophysical behavior of materials is governed by their atomic level excitations and how they store heat. Most of the thermal energy excites oscillations of the atoms, quantized as phonons, but in magnetic materials a considerable amount of heat is also absorbed by fluctuations of the electronic spins. This thesis explores the thermodynamics of phonons and magnetic spins in Fe-systems: we investigate the coupling between these excitations in Fe, Fe-Ni, and Fe-C, quantify their size dependency in nanocrystalline in Ni₃Fe, and assess their individual roles in the anomalous thermal expansion of Fe-Ni Invar.
Most materials expand when heated due to enhanced atomic oscillations. However, in 1895 C.E. Guillaume combined Fe and Ni to discover a material with near-zero thermal expansion, called Invar. This discovery was awarded the 1920 Physics Nobel Prize and sparked thousands of scientific investigations. Since the anomalous Invar effect is associated with magnetism, nearly all studies have focused on the electronic and spin structure of Fe-Ni. But phonons are needed to complete the picture, and to date, the anomalous Invar behavior is not fully understood. Here, we explore a method for measuring thermal expansion that is capable of isolating contributions from phonons and spins. Since the thermal energy of materials is related to entropy, the thermal expansion can be indirectly determined through individual entropic contributions by using a Maxwell relation. The phonon and magnetic entropies were measured by combining two nuclear resonant x-ray scattering techniques, with samples under pressure in diamond-anvil cells. We show that the Invar behavior stems from a competition between phonons and spins, that oppose each other for near-zero thermal expansion. A spin-phonon coupling improves the precision of this cancellation, extending the range of Invar behavior.
Such a coupling of phonons and spin was also observed in pure Fe and Fe₃C cementite, as their phonon energies correlate to the change in magnetization. This motivated us to develop a magnetic quasi-harmonic model for Fe and Fe₃C, which accounts well for the deviation of phonon energies from pure volumetric effects of the conventional quasi-harmonic approximation.
The thermodynamics of materials is also affected by the size of their crystallites. We determined the size effects on the heat absorption by phonons, electrons, and spins in nanocrystalline Ni₃Fe. All excitations become enhanced in the nanomaterial. In particular, the redistribution of spectral weights amplifies the phonon entropy. This helps stabilize the nanostructure against the enthalpy from its extra grain boundaries. However, the nanostructure is meta-stable, and the grains will grow into their bulk counterpart when diffusion is enabled at elevated temperatures.</p
Geometry and Dynamical Systems in Machine Learning and Control
Get PDFFor many problems of interest in machine learning and control, we have access to rich information about underlying geometry and dynamics; we can leverage this information to build robust and performant solutions in new algorithms, optimizations, and designs. In this thesis we study four problem settings to stress this central assumption. First, we study conformal generative modeling, using computational geometry techniques to simplify and register complex 2D surfaces and enabling the use of a variety of flow-based generative models as plug-and-play subroutines. Second, we study data-driven robust optimization problems in control, modeling the precise impact of dynamics uncertainty in several control frameworks using convex geometry. Third, we study compactly-restrictable policy optimization, constraining the available states and actions in reinforcement learning and optimal control problems to be consistent with the inherent dynamics of the systems to be controlled. Finally, we study nonlinear model predictive control on Lie groups as applied to a 3D hopping robot platform, developing a control methodology compatible with nontrivial state space geometry and hybrid system dynamics