Cornell Theses and Dissertations
Permanent URI for this collection
The theses and dissertations of graduate students at Cornell University have been deposited in Cornell's institutional repository (eCommons) since about 2004. This collection also includes a few earlier Cornell theses.
Students retain ownership of the copyright of their work. Students also have the option of imposing a temporary embargo on access to the full text of their theses for limited amount of time (see eCommons access policy). If access to a thesis is restricted, the metadata record for the thesis is still visible, but the text "Access to Document Restricted" is displayed, and a field labeled "No Access Until," which indicates the date when the full text of the thesis will become accessible.
More information about finding Cornell theses and dissertations is available on this library guide, and the eCommons help page for finding content in specific collections, including theses and dissertations.
In general, older theses and dissertations from Cornell University are not currently available as digital files in eCommons. The Library is willing to digitize and make available older Cornell theses on a cost recovery basis. If you are interested in this service, please contact dcaps@cornell.edu.
Browse
Recent Submissions
Item On the p-class groups of the pure number field Q(N^(1/p)) and its Galois closure Q(N^(1/p), zeta_p)Murgescu, Radu Andrei (2009-12)We study the class groups of the fields $K$ = $\mathbb{Q}(N^{1/p})$ and $M$= $\mathbb{Q}(N^{1/p}, ζ_p)$, where $N$ and $p$ are primes and $ζ_p$ is a primitive $p^{th}$ root of unity. Furthermore, we restrict ourselves to the study of the $p$-class groups, i.e. the Sylow $p$-subgroups of the class groups in question. We denote the $p$-class group of a field $F$ by $S_F$ . Define $rank(S_F)$ as the dimension of $S_F/(S_F)^p$ as a vector space over $\mathbb{F}_p$. $\\$ Frank Gerth III, in [9], settled a problem left open by F. Calegari and M. Emerton in [4]. He, in turn, posed a related question, which we answer in this thesis, as follows. Theorem 3.15. Let $N$ ≡ 4 or 7 (mod 9), and suppose 3 is a cubic residue (mod $N$). Then $rank(S_M)$ = 2. For general values of $p$, we obtain results (Propositions 2.13, 4.1, 4.3, 5.2, 5.3, and 5.9) on the existence of certain norms in the extension $M/\mathbb{Q}(ζ_p)$, and use these results to give bounds on $rank(S_M)$. For example, we show the following: Proposition 5.2. Let $p$ and $N$ be any primes with $p$ ≥ 5, $N \neq p$, and let $f$ denote the minimal positive integer $x$ such that $N^x$ ≡ 1 (mod $p$). Let ${U_{\mathbb{Q}(ζp)}}^+$ be the real units of $\mathbb{Q}(ζ_p)$. Suppose $\frac{p-1}{f}$ is odd. Then ${U_{Q(ζ_p)}}^+ \subseteq N_{M/\mathbb{Q}(ζ_p)}(M^\ast)$, where $N_{M/\mathbb{Q}(ζ_p)}$ denotes the usual norm map of the extension $M/\mathbb{Q}(ζ_p)$.
Item MULTI-FUNCTIONAL MODULAR BIO-INSPIRED UNDULATING ROBOTLiu, Tianbin (2025-05)Microplastics in natural aquatic environments have caused major environmental crises due to their resistance to biodegradation and ingestion toxicity. Conventional cleaning methods, such as tow nets and propeller-based suction pumps, are inadequate due to their limited maneuverability and high levels of biological intrusion. In this study, we present a multifunctional bio-inspired robotic module, inspired by the efficient filter-feeding mechanism of the Giant Larvacean, for active particle-laden fluid suction. The design is fundamentally based on the fluid–structure interaction of a flexible undulating structure, employing non-uniform stiffness distribution and a tailored actuation scheme to optimize the trade-off between suction and propulsion. We analyze single-module performance at various actuation frequencies, with particular attention to the effect of the ratio between the natural frequency and the input frequency on pumping and propulsion, and identified optimal motions based on considerations of efficiency, stability, and environmental compatibility. Furthermore, we designed the robot structure based on the Giant Larvacean mechanism, and proposed a FastPID algorithm to reduce the tuning time required in systems that do not demand high-precision control. Lastly, we demonstrate the feasibility, adaptability, and maneuverability of multimodule assemblies. This flexible bio-inspired robotic module opens up a novel pathway toward low-impact microplastic remediation with the potential for large-scale field deployment.Item AN APPLICATION OF THE TRIANGULAR THEORY OF LOVE TO ONE’S LOVE OF THEIR ACADEMIC CONCENTRATIONHayes, Aurora (2025-05)This study sought to investigate whether the Triangular Theory of Love could be applied to the love an individual may feel toward their academic concentration in higher education.While previous studies have focused on loving relationships in romantic, sexual, and parental contexts, this study, to the best of our knowledge, is the first to attempt to apply a theory of love in an educational context to a student’s academic concentration. In an adapted version of the Triangular Love Scale and replication of the previous methodology used in The Love of one’s musical instrument as a predictor of happiness and satisfaction with musical experience, this study sampled undergraduate psychology majors at an elite university in the northeastern United States to gauge their emotional, motivational, and cognitive disposition toward their academic concentration (Sternberg, 2023). Results of this study showed that an application of an adapted version of the Triangular Love Scale yields statistically significant and reliable results. Each component of the Triangular Theory of Love (Commitment, Passion, and Intimacy) was measured with high internal consistency reliability, and the principal component analysis of thesubscales loaded distinctly to each construct, with each accounting for 33.33% of the variance within the population. In light of these findings, we believe this research provides the preliminary steps in the continued validation of applying the Triangular Theory of Love to measure student’s affinities toward their academic concentration.Item Essays on Digital Platforms: Platform Design and Impact of DigitilizationZuo, Si (2025-05)Digital platforms have fundamentally transformed economic interactions by reducing transaction costs, facilitating information exchange, and shaping market structures. These platforms rely on algorithms, reputation systems, and pricing mechanisms to create efficient marketplaces. However, the design choices of these platforms - such as rating granularity, pricing strategies, and algorithmic personalization - can have profound implications for fairness, efficiency, and long-term market outcomes. At the same time, the rapid digitization of commerce has altered the dynamics of online and offline interactions. While digital platforms provide opportunities for businesses to expand their reach, they also introduce challenges related to market cannibalization and algorithmic dependence. This dissertation examines platform design and the impact of digitization in digital markets.In Chapter 2, I study rating granularity on online platforms by developing a dynamic model that reveals a fairness-efficiency trade-off. Coarser ratings support new entrants but lead to information loss, and I identify an optimal granularity level that maximizes platform revenue. Chapter 3 explores price signaling in service platforms, where I provide empirical evidence that high-quality firms use low introductory prices to build reputation, highlighting the role of pricing in quality discovery. Chapter 4 investigates algorithmic dependence in personalized recommendation systems, demonstrating that while personalization enhances short-term user welfare, it hinders long-term preference learning, with policy interventions proposed to balance these effects. Chapter 5 analyzes the impact of online store openings on brick-and-mortar businesses, showing that product categories react differently to digital expansion, offering new insights into the dynamics of omnichannel retail and providing actionable guidance for adapting mall strategy in the digital era. This dissertation contributes to platform economics, algorithmic personalization, and digital commerce by providing theoretical insights and empirical evidence to guide platform design and business strategy.Item THE INS AND OUTS OF MANGANESE: ROLES OF ZIP14 IN BLOOD-BRAIN BARRIER TRANSPORT AND ASTROCYTE FUNCTIONZou, Jiaqi (2025-05)Manganese (Mn) is essential for many enzymatic processes in the brain; however, Mn overload can lead to neurotoxicity and behavioral deficits. The metal transporter ZIP14 (SLC39A14) plays a key role in systemic Mn homeostasis, yet its role in brain Mn homeostasis remains unclear. This dissertation investigates ZIP14’s roles in brain endothelial cells and astrocytes.The blood-brain barrier (BBB), comprised of polarized endothelial cells, tightly regulates metals in and out of the brain. However, ZIP14’s precise localization and role in BBB endothelial cells have yet to be defined. Here, we show in vivo ZIP14 expression in BBB endothelial cells, which upregulates following Mn supplementation. Expansion microscopy revealed a shift in ZIP14 localization from equal apical-basolateral distribution to predominantly basolateral after Mn exposure. Endothelial-specific Zip14 KO (EKO) mice exhibited impaired Mn efflux from the brain and increased brain Mn accumulation after Mn supplementation. In vitro studies using EKO-derived primary endothelial cells and ZIP14-overexpressing hCMEC/D3 cells confirmed that ZIP14 facilitates basolateral-to-apical Mn transport. These results demonstrate that ZIP14 is critical for brain Mn clearance, We also observed ZIP14 expression in primary astrocytes isolated from wild-type mice. Using astrocytes derived from astrocyte-specific Zip14 knockout mice, we demonstrated that ZIP14 mediates Mn uptake into astrocytes. Additionally, ZIP14 was partially localized to the mitochondria, and its deletion in astrocytes protected against the Mn-induced mitochondrial dysfunction, as evidenced by the preserved mitochondrial morphology, improved mitochondrial respiration and metabolic activity compared to controls. Collectively, these studies establish ZIP14 as a key mediator of brain Mn homeostasis, with cell-specific roles: promoting Mn clearance from the BBB and contributing to Mn toxicity in astrocytes. Our findings advance the understanding of ZIP14’s functions in Mn neurotoxicity mechanisms and may serve as a potential therapeutic target to mitigate Mn-induced neurotoxicity.Item Next Generation Spectrometers for Far-infrared/Submillimeter AstrophysicsZou, Bugao (2025-05)Far-infrared (FIR) and submillimeter astronomy is a crucial frontier in modern astrophysics, providing unique insights into the formation and evolution of cosmic structures that remain hidden at other wavelengths. This thesis presents the development of advanced spectroscopic instrumentation designed to push the boundaries of observational capabilities in FIR/Submm astronomy. By addressing key limitations in current technology, these instruments enable groundbreaking studies of galaxy formation, planetary system evolution, and the chemical composition of astrophysical environments. At the core of this work is the development of high-performance spectrometers for two state-of-the-art observatories: the Epoch of Reionization Spectrometer (EoR-Spec) to be used on the Fred Young Submillimeter Telescope (FYST), a 6 meter aperture telescope sited at an elevation of 5600 meters on Cerro Chajnantor, and the Virtually Imaged Phased Array (VIPA) on the Planetary Origins and Evolution Multispectral Monochromator (POEMM), a NASA mission on with a 1.8 meter telescope suspended at 35 km elevation on a long duration balloon. EoR-Spec is designed to study star and galaxy formation and the growth of structure from the epoch of reionization, when the first luminous sources began to ionized the neutral intergalactic medium to cosmic noon when the peak of star formation per unit co-moving volume occurred. At the high site, FYST leverages exceptionally dry atmospheric conditions to conduct sensitive measurements in the millimeter and submillimeter regimes. POEMM will explore the complex chemistry of protoplanetary disks and planetary atmospheres, offering critical insights into the processes that shape planetary systems. This thesis bridges technological innovation with scientific discovery, demonstrating how novel spectrometer designs can enhance instrument performance and expand the reach of FIR/Submm astronomy. By overcoming challenges related to spectral resolution and detector sensitivity, the instruments presented here open new pathways for investigating the Universe’s most elusive phenomena. The advancements documented in this work not only enhance current observational capabilities but also lay the groundwork for future missions that will continue to redefine our understanding of the cosmos.Item IN VIVO IMAGING OF CEREBRAL BLOOD FLOW AND NEURAL ACTIVITY CHANGES IN THE CONTEXTS OF ALZHEIMER’S DISEASE AND PSYCHEDELICSZirkel, Rick (2025-05)The brain is a highly dynamic organ in which neural activity, vascular regulation, and molecular signaling are intricately intertwined. Understanding these relationships is essential for uncovering the mechanisms that underlie both cognitive function and dysfunction. In vivo imaging enables the direct measurement of neural activity and cerebral blood flow in living animals with cellular resolution, providing a powerful window into brain function in both health and disease (chapter 1). This dissertation leverages a range of optical imaging tools to examine how cerebral blood flow and neural dynamics, and the relationship between them, are impacted in two distinct contexts: Alzheimer’s disease (AD) (chapter 4) and psychedelic drug action (chapter 5). Alzheimer’s disease is characterized by disruptions in both neural activity and blood flow, yet the precise mechanisms through which these impairments arise—and whether they can be reversed—remain unclear. In the first focus of this thesis, I review our understanding of these impairments, as well as current research examining therapeutic targets to eliminate such dysfunction (chapter 2). Next, using in vivo two-photon microscopy, I demonstrate how a specific impaired activity pattern can be improved after the restoration of cerebral blood flow (CBF) deficits. Increasing CBF in AD model mice leads to enhanced orientation tuning in visual cortex neurons, providing a potential therapeutic mechanism for restoring cortical function (chapter 4). For the second focus of this thesis, I leverage in vivo imaging techniques to evaluate cerebral blood flow and neural activity changes in the context of psychedelics, specifically psilocybin. Psychedelics are known to profoundly alter consciousness and have promising therapeutic potential for treating mood disorders such as anxiety, depression, and addiction (chapter 3). However, their effects on the coordination between neural and vascular activity in the brain, a relationship known as neurovascular coupling (NVC), are less well understood. Clinical fMRI studies investigating psychedelic drug action rely on NVC to examine blood flow changes as a proxy for neural activity. Using micro- and mesoscale in vivo imaging techniques, I show that psilocybin alters stimulus-evoked neurovascular coupling in mouse visual cortex and discuss potential mechanisms underlying these changes, and how these alterations may influence the interpretation of fMRI imaging with psychedelics (chapter 5). Together, these studies highlight the power of in vivo imaging to uncover complex, dynamic interactions between neural activity and cerebral blood flow across multiple spatial and temporal scales. This dissertation not only demonstrates the utility of these imaging techniques in identifying neural and vascular impairments in AD and psychedelic drug action but also provides a framework for how these tools can be used to guide future therapeutic strategies. By combining cellular-resolution imaging with systems-level analysis, this work contributes to a deeper understanding of brain function in both pathological and pharmacologically altered states (chapter 6).Item TIME-RESOLVED X-RAY STRUCTURAL BIOLOGY WITH MICROFLUIDIC MIXERSZielinski, Kara (2025-05)X-ray based structural biology has advanced the understanding of biological macromolecules, such as proteins and nucleic acids, by providing atomic level detail. Biological macromolecules, however, are not static and have intricate interactions with many binding partners, including ions, small molecules, and other macromolecules. Recent advances in X-ray sources, with the development of X-ray Free Electron Lasers and fourth-generation synchrotrons, have contributed to the exciting development of time-resolved techniques that allow for structural dynamics to be directly measured. In particular, microfluidic mixers are extremely versatile and well posed to rapidly combine multiple species for reaction initiation. The freshly mixed sample can then be directly delivered, at pre-set time delays, to the X-ray beam to probe its structure as the reaction progresses. Here, I present several successful applications of time-resolved X-ray experiments with diverse biological samples, various structural biology techniques, multiple X-ray sources, and different types of microfluidic mixers. These developments expand the types of time-resolved experiments that can be performed, and overcome previous limitations related to X-ray source type and the size of the species that can be mixed.Item OPERANDO MICROSCOPY FOR MECHANISTIC INSIGHTS INTO COCATALYST-MODIFIED SEMCONDUCTOR PHOTO(ELECTRO)CATALYSISZhao, Zhiheng (2025-05)This dissertation presents studies on semiconductor photoelectrochemical systems, with 2 chapters detailing my major contributions and 3 appendices covering collaborative efforts with my minor contributions.Chapter 1 investigates how single cocatalyst particle depositions influence local and remote charge dynamics on particulate BiVO4 photoanodes. I led the experimental design, synthesis, single-molecule fluorescence imaging, and analysis. Results show that single cocatalyst particles enhance local catalytic activity, while remote suppression occurs for cocatalyst CoOx deposition on basal facets. Operando single-molecule imaging reveals synergistic effects for cocatalyst CoOx and anti-synergistic effects for cocatalyst Pt/PtOx, contributing to improved photocatalyst performance. This chapter comes from a completed first-authored manuscript. Chapter 2 explores overall water splitting on InGaN/GaN nanowires with integrated cocatalysts. I applied operando fluorescence microscopy to map electron- and hole-induced reactions by tracking resazurin reduction and amplex red oxidation. Imaging revealed reaction "hotspots" where electrons and holes converge at shared HER and OER cocatalyst sites, validating critical design principles for optimized nanowires. This chapter comes from a completed manuscript on which I am a co-first author. Appendix 1 examines long-range micropollutant adsorption on metal-promoted photocatalysts (Nature Catal. 2024, 7, 912–920). Appendix 2 investigates nanoscale cooperative adsorption for materials control (Nature Commun. 2021, 12, 4287). Appendix 3 uncovers energy conversion pathways in biohybrids using single-cell multimodal imaging (Nature Chem. 2023, 15, 1400–1407). This dissertation demonstrates the potential of operando single-molecule and single-particle microscopy for elucidating cocatalyst-induced mechanisms, advancing both photo(electro)chemical and biohybrid catalytic system design.Item Reasoning in the WildZhao, Wenting (2025-05)Teaching machines to reason has been a longstanding goal in artificial intelligence. Recently, the rapid advancement of language modeling has advanced this vision, opening up new possibilities for automated reasoning. Although existing benchmarks have demonstrated strong reasoning performance of LMs, it remains unclear how effectively such models reason in real-world scenarios, where queries differ significantly in complexity and style from the standard evaluation datasets. This dissertation identifies two main obstacles that prevent LLMs from reasoning effectively in realistic settings: (1) distributional mismatches between standard training data and the user queries encountered in the wild, and (2) the difficulty and cost associated with collecting expert-annotated training data for complex reasoning tasks. To better assess reasoning performance under realistic conditions, we first introduce data sources and evaluation benchmarks directly collected from real-world use cases, establishing representative real-world reasoning challenges. Analysis of these benchmarks reveals significant limitations in contemporary language models, highlighting areas that require progress. Subsequently, we explore training approaches using alternative supervision that enable reasoning without reliance on manually annotated data. We investigate structural supervision, an approach that incorporates prior knowledge about the underlying structure of reasoning tasks into latent variable models, enabling them to better handle different reasoning scenarios, such as multi-hop inference and abductive reasoning. Additionally, we explore using language agents for complex reasoning tasks. Language agents utilize environmental feedback, where they learn iteratively by interacting with an external environment rather than from explicit annotations.