University of Illinois Urbana-Champaign
Illinois Digital Environment for Access to Learning and Scholarship RepositoryNot a member yet
123813 research outputs found
Sort by
Essays on Institutions and Inequality
No full textSubmission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2027-08-01The student, Alok Ranjan, accepted the attached license on 2025-07-07 at 10:46.The student, Alok Ranjan, submitted this Dissertation for approval on 2025-07-07 at 10:55.This Dissertation was approved for publication on 2025-07-07 at 16:39.DSpace SAF Submission Ingestion Package generated from Vireo submission #22421 on 2025-10-25 at 15:53:06This dissertation, titled Essays on Institutions and Inequality, brings together three empirical studies that examine how institutional rules and design features shape economic behavior and distributive outcomes. Using large-scale administrative and survey data, each chapter employs quasi-experimental methods to identify causal effects in distinct domains: judicial decision-making, labor market dynamics, and political participation. The first chapter, Judicial Transparency and Criminal Justice (with Felipe Diaz and Anna Kyriazis), investigates how greater public visibility affects judges’ discretionary decisions in criminal courts. Elected judges wield considerable power over monetary sanctions, yet their choices are often shielded from public scrutiny. At the same time, local governments increasingly rely on court-imposed legal financial obligations (LFOs)—fines, fees, and surcharges that disproportionately burden low-income defendants and often yield limited net revenue once enforcement costs are considered. This chapter examines whether transparency can realign judicial behavior with public accountability and distributive goals. Leveraging a policy reform in North Carolina that mandated public reporting of court cost waivers, we employ a regression discontinuity in time design to compare outcomes for defendants sentenced immediately before and after the reform. The findings show that public reporting reduces the likelihood that judges impose court costs, improves the targeting of waivers toward economically vulnerable defendants, and—counterintuitively—may increase total collections by enhancing payment compliance. The results highlight how transparency, under the right institutional conditions, can improve both equity and fiscal efficiency in the criminal justice system. The second chapter, Labor Market Dynamics of New Mothers (with Mark Borgschulte and Rebecca Thornton), explores the short-run and medium-run labor market responses to childbirth. Childbirth marks a pivotal turning point in women's labor market trajectories, yet most data sources obscure when and how these disruptions begin. Using restricted-access data from the American Community Survey (ACS) linked to the Census Household Composition Key (CHCK), we construct a novel event-time dataset that tracks parents’ employment patterns month-by-month over a 48-month window centered on childbirth (–24 to +24 months). We document that maternal employment begins declining as early as six months before birth, with nearly 60\% of the total decline occurring prior to delivery—a “pregnancy penalty” that precedes most postpartum-focused policy interventions. These early exits are steepest among non-White and non-college-educated mothers, but these groups also exhibit faster post-birth recovery, in contrast to the slower rebound observed for White and college-educated mothers. Fathers’ labor market outcomes, by contrast, remain stable across this same window. We then examine whether workplace breastfeeding accommodations introduced under the Affordable Care Act (ACA) improved maternal labor supply. Exploiting variation in prior state-level mandates, we find no evidence that the ACA provisions affected mothers’ employment or timing of return to work. The third chapter, Jurisdiction Size and Political Participation, examines how the size of local political jurisdictions influences voter turnout and candidate selection. I leverage a quasi-natural experiment in the Indian state of Uttar Pradesh, where a population threshold rule exogenously determined the size of over 50,000 local government units. Using a regression discontinuity design, I find that a 10% increase in jurisdiction size leads to a 0.6% decline in voter turnout in local elections. While the number of contesting candidates increases with electorate size, the quality of candidates and the characteristics of winners do not vary systematically. I also test the rational choice model of voting by simulating millions of elections and show that voter turnout increases with the probability of being pivotal. Together, these chapters offer new insights into how institutional features influence individual behavior and shape economic and civic outcomes. The analysis spans transparency mandates in criminal courts, regulatory protections in the labor market, and the structure of democratic governance. By identifying how public institutions affect decisions related to justice, employment, and participation, the dissertation contributes to policy debates on institutional accountability, labor market policy, and democratic engagement
Understanding molecular interactions and design of selective electrochemical processes for water treatment
No full textSubmission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2027-08-01The student, Anaira Roman Santiago, accepted the attached license on 2025-07-07 at 10:16.The student, Anaira Roman Santiago, submitted this Dissertation for approval on 2025-07-07 at 16:34.This Dissertation was approved for publication on 2025-07-08 at 15:03.DSpace SAF Submission Ingestion Package generated from Vireo submission #22425 on 2025-10-25 at 15:53:07Water scarcity, resource depletion, and exposure to anthropogenic contaminants have been major drivers for the development of efficient technologies for water remediation. Recently electrochemical separations have emerged as a promising solution due to their modularity, scalability, and potential for net-zero emissions when powered by renewable energy. However, their broader adaptation is limited by poor ion selectivity, which increases energy demands. This challenge is largely dependent on the electrode materials used in electrochemical systems. Developing tailored materials with molecular-level selectivity can help achieve energy-efficient separations. For my doctoral thesis, I investigated the molecular-level interactions and interface morphology governing electrochemical separations of per- and polyfluoroalkyl substances (PFAS), phosphates, and rare earth elements (REE) from water, aiming to enhance selectivity and efficiency in pollutant removal and resource recovery. By designing and characterizing redox-active polymers and ligands, we identified key interactions─ electrostatic, redox, fluorophilic and non-covalent interactions—that drive selective binding and release of target species. For PFAS, we demonstrate that electrostatic interactions predominantly govern capture, while fluorophilic groups selectively enhance uptake of short-chain PFAS. Polymer structure and porosity, influenced by fluorinated monomer content, were shown to further modulate PFAS selectivity via steric effects. Furthermore, integration of electrosorption with electrodegradation enabled energy-efficient PFAS destruction by concentrating contaminants in wastewater prior to oxidation. To recover phosphorus, we developed ferrocene-based polymers which achieved high selectivity over competing ions such as chloride. In-situ neutron reflectometry revealed solvation-driven ion selectivity under dynamic redox modulation, providing insights into polymer design and separation treatment trains. For REE extraction, we demonstrated how the design of ligands can be accelerated via integration with artificial intelligence and chemical language models to yield selective chemistries that lower separation costs. Across all systems, the interplay of interfacial morphology, hydration, and chemical functionality was shown to be critical for efficient separations. The findings of this dissertation provide a framework for designing advanced interfaces for selective separations, with broader implications for sustainable water treatment and applications in industries requiring precise molecular selectivity
Teacher self-efficacy for classroom management, behavior management practices, and years of experience: factors of influence on teacher well-being
No full textSubmission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2027-08-01The student, Jenna Mahoney, accepted the attached license on 2025-07-10 at 11:17.The student, Jenna Mahoney, submitted this Dissertation for approval on 2025-07-10 at 11:18.This Dissertation was approved for publication on 2025-07-13 at 07:59.DSpace SAF Submission Ingestion Package generated from Vireo submission #22455 on 2025-10-25 at 15:53:23Teachers in the United States report the lowest levels of well-being among helping professionals, a trend exacerbated by the COVID-19 pandemic and the increasing demands placed on educators. Burnout—heightened to unprecedented levels during and after the pandemic—remains a significant contributor to reduced teacher well-being (TWB) and rising attrition rates. A major source of teacher burnout is the stress associated with managing student behavior. Understanding how teacher self-efficacy for classroom management (TSECM) and behavior management practices (BMP) relate to TWB is therefore critical, particularly across different stages of teaching experience. TSECM refers to a teacher’s belief in their ability to effectively manage classroom dynamics, while BMP encompasses the strategies employed to prevent and address behavioral challenges. Novice teachers, who often report higher stress levels, may do so due to lower self-efficacy and limited familiarity with effective BMPs. This study investigates the roles of TSECM, BMP, and years of teaching experience in influencing TWB. Cross-sectional survey data were collected from an urban school district in a small Midwestern city, from 189 classroom teachers in January 2024. Participants completed validated measures of TSECM, BMP, and TWB, along with demographic data including years of experience. The study addressed three research questions: (1) Do years of teaching experience moderate the relationships between TSECM, BMP, and TWB? (2) What are the moderating effects of TSECM and BMP on TWB in the context of experience? (3) Do TSECM and BMP mediate the relationship between years of experience and TWB? Findings from moderation and mediation analyses were mixed. A significant interaction was found between TSECM and BMP in the association with TWB (B = .313, SE = .073, p < .001), indicating that the combination of high TSECM and high use of BMP is associated with greater teacher well-being. However, other moderation effects involving years of experience were not significant, nor were the mediation effects of TSECM and BMP in the relationship between experience and TWB. Considering that years of experience have not been identified as having a significant association with TWB, whereas the interaction between TSECM and BMP has proven to be influential, findings that educators across all stages of their careers would gain from enhanced support in BMPs and initiatives aimed at enhancing TSECM. These findings carry three key implications. First, they highlight the complexity of years of teaching experience as having an association with well-being. Second, they underscore the potential of TSECM and BMP as malleable constructs with meaningful effects on TWB. Third, they contribute to a deeper understanding of how these constructs interact, offering direction for future research and professional development focused on enhancing teacher well-being through targeted support in classroom management and self-efficacy building
Design and analysis of a compact high-speed air-breathing engine
No full textSubmission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2027-08-01The student, Nozomu Kato, accepted the attached license on 2025-07-09 at 14:50.The student, Nozomu Kato, submitted this Dissertation for approval on 2025-07-09 at 15:18.This Dissertation was approved for publication on 2025-07-10 at 11:28.DSpace SAF Submission Ingestion Package generated from Vireo submission #22466 on 2025-10-25 at 15:53:24This dissertation investigates the underlying physics of challenges associated with compact, high-speed, air-breathing propulsion systems and examines potential solutions. The primary focus is on the unique technical challenges posed by the stringent requirement for a compact design, which will fundamentally alter system performance. Such propulsion systems are critical for the development of platforms such as the Air-Launched Effect (ALE)—a small, unmanned reconnaissance aircraft. To address these challenges, this study combines component design, theoretical analysis, high-speed wind tunnel experiments, and reduced-order modeling. Component-level analyses revealed critical design trade-offs. For the inlet, it was found that while scoop inlets offered high theoretical efficiency, truncated planar two-dimensional inlets provided superior performance for compact designs. For the combustor, experiments demonstrated that newly designed inserts, particularly mesh-based inserts, were capable of achieving stable and more compact combustion compared to a baseline configuration. For the exhaust nozzle, analysis showed that designing for under-expanded conditions enabled a remarkable 44 % length reduction with a corresponding thrust penalty of 20 %. Subsequently, these component-level findings were integrated into a reduced-order model. After its combustor sub-model was tuned and validated against the experimental data, the complete model was used to produce a preliminary integrated engine design, 1319.2 mm in length, capable of generating 137.8 N of thrust for the target flight condition. The primary contribution of this work is a complete design-and-analysis framework that provides critical insights into the performance of compact ramjet/scramjet components and delivers a validated tool for the rapid preliminary design of such propulsion systems
Forming modern times: the transatlantic day novel, 1884-1941
No full textSubmission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2027-08-01The student, Aidan Watson-Morris, accepted the attached license on 2025-07-09 at 17:39.The student, Aidan Watson-Morris, submitted this Dissertation for approval on 2025-07-09 at 17:48.This Dissertation was approved for publication on 2025-07-10 at 14:30.DSpace SAF Submission Ingestion Package generated from Vireo submission #22472 on 2025-10-25 at 15:53:26Between the years of 1880 and 1940, an increasingly popular feature of the anglophone novel—its restriction of narrative duration to a single day—became paradigmatic of its era. Forming Modern Times: The Transatlantic Day Novel, 1884-1941 narrates transatlantic literary history with this idiosyncratic form as the protagonist: the single-day novel. The single-day novel locates a cross section of literary history that overlaps with the modernist canon but also finds once popular, now little-read texts in the fault lines of the archive. I tell the history of the day novel without its best-known modernist exemplars, James Joyce’s Ulysses (1922) and Virginia Woolf’s Mrs. Dalloway (1925), to recover a microtradition at its moment of becoming. These texts speak to each other in a frequency only audible through an attunement to the formal logics of the single day novel in the first half of the twentieth century, when writers are thinking through the portability and problems of a literary paradigm still in the making. The novel of the single day asks its readers to question the forms of time that pace out the rhythms of social life, especially as mediated by the abstract structures and strictures of work in global capitalism. A cultural desire for a “universal day” or “universalized everyday,” manifest in the 1884 Prime Meridian Conferences, conditions the utopian dream of everyday aesthetics at the turn of the century. As the single day became a figure for collective time, such novels offered a staging ground for the remaking of social time. Novels such as Frank Swinnerton’s Nocturne (1917), Nathalie Colby’s Black Stream (1927), John Rodker’s Adolphe 1920 (1929), Storm Jameson’s A Day Off (1933), Betty Miller as B. Bergson Spiro’s The Mere Living (1933), Henry Green’s Party Going (1939), Richard Wright’s Lawd Today! (completed in 1937, published in 1963), and Virginia Woolf’s Between the Acts (1941) all fissure official time and the time of everyday life. Developing a minimal teleology, I organize these novels into the decades of their publication, as the form pursues its routes through avant-garde, political, and commercial traditions. Following the aleatory trajectory of the form as it runs orthogonal to conventional accounts of modernism and traverses competing value systems through the first half of the century, I map a strange path through the archive that opens onto broader questions of aesthetic form, labor and leisure, and the temporality of everyday life—questions that authors rephrase, re-inflect, and reanimate to diverse ends. In their form, these novels produce a theory of the synecdochic relations between part and whole that informs my method of constellating each day novel as a moment of literary-historical time. This literary history in intervals provides a prehistory of the present, as an all but unnoticed twenty-first-century day novel renaissance mirrors the early twentieth century, conditioned in the contemporary period by a parallel, radical intrusion of work into the time of daily life. To read the day novel today aslant the day novel of the twentieth century renews a collective dream of the different forms our everyday lives might assume
Metabolomic insights into Pseudomonas aeruginosa behavior, secretome, and biofilm development using mass spectrometry imaging
No full textSubmission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2027-08-01The student, Joenisse Rosado-Rosa, accepted the attached license on 2025-07-11 at 10:37.The student, Joenisse Rosado-Rosa, submitted this Dissertation for approval on 2025-07-11 at 10:46.This Dissertation was approved for publication on 2025-07-15 at 16:37.DSpace SAF Submission Ingestion Package generated from Vireo submission #22505 on 2025-10-25 at 15:53:46Since the early twentieth century, antibiotics have been employed to treat bacterial infections, but bacteria have constantly evolved mechanisms that combat antibiotic treatment. These evolved mechanisms, including biofilm formation and antibiotic-inactivating enzymes, have been studied as potential factors for antibiotic resistance. Pseudomonas aeruginosa, also selected to be one of the multidrug-resistant ESKAPE pathogens, is a Gram-negative nosocomial bacterium responsible for about 7% of all hospital-acquired infections. P. aeruginosa is often referred to as an opportunistic pathogen, known to infect patients with low immunity, such as those who have been diagnosed with cystic fibrosis (CF), cancer, and other immunocompromising diseases. P. aeruginosa is the model organism for biofilm studies due to its robust understanding of the production of biofilm. During biofilm production, P. aeruginosa secretes secondary metabolites into the biofilm matrix. While the characterization of some of these molecules has increased our understanding of biofilm initiation, production, and maturation processes, some key aspects of these processes and secondary metabolite functions, such as some environmental conditions required for secretion and the effects of external small molecule stressors, have yet to be elucidated. To illuminate these understudied factors, my work explores the impact of small-molecule perturbations on the growth, biofilm production, and secretion profile of three molecular types. Specifically, the impact D-amino acids, bio-metals, and rhamnolipid congeners are explored on the growth, biofilm production, and secretion profile of P. aeruginosa in vitro, ex vivo lung tissues, and abiotic surfaces. These studies used mass spectrometry imaging (MSI), a chemical imaging technique used for its functionality in studying the spatial distribution of molecules on a two-dimensional surface. This technique is a label-free method suitable for biological samples due to its ability to analyze complex samples with high sensitivity and has been commonly used to study biofilm models on a variety of bacterial samples. An improved understanding of the effects these small molecules have on the pathogenicity and biofilm development of P. aeruginosa may have broader implications in developing treatments for infection by biofilm-producing organisms
Computational modeling of test articles in the PlasmatronX inductively coupled plasma facility
No full textSubmission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2027-08-01The student, Abhyudaya Singh, accepted the attached license on 2025-07-17 at 16:03.The student, Abhyudaya Singh, submitted this Thesis for approval on 2025-07-17 at 16:15.This Thesis was approved for publication on 2025-07-18 at 14:44.DSpace SAF Submission Ingestion Package generated from Vireo submission #22605 on 2025-10-25 at 15:54:13This thesis presents a detailed numerical investigation of plasma–material interactions under non-local thermodynamic equilibrium (NLTE) conditions, using the PlasmatronX inductively coupled plasma (ICP) facility at the University of Illinois Urbana-Champaign as the reference testbed. The study is motivated by the need for accurate prediction of surface heat flux and species behavior in high-enthalpy environments relevant to atmospheric reentry and thermal protection system (TPS) design. A multi-physics simulation framework was employed, coupling a finite-volume NLTE flow solver (HEGEL), a finite-element electromagnetic solver (FLUX), and a detailed thermochemical and transport property library (PLATO). The framework accounts for multi-temperature thermochemistry, electromagnetic power deposition, and finite-rate gas–surface interactions. Three complementary studies were performed. First, axisymmetric NLTE simulations of a calorimetric probe (isoQ30) were conducted across varying RF(Radio Frequency) power and chamber pressures. A stagnation-line boundary layer formulation was used to estimate wall catalytic activity, and the computed heat fluxes and nozzle exit enthalpies showed strong agreement with experimental measurements. Second, simulations at 55~kW and 200~mbar were validated against TALIF-based profiles of temperature and atomic species, demonstrating accurate reproduction of experimental trends. Third, two-dimensional and three-dimensional simulations over a graphite wedge test article were carried out, incorporating finite-rate gas–surface reactions. These analyses revealed significant production of carbonaceous species and highlighted the role of lateral spreading, vortex roll-up, and compositional mixing in shaping the heat flux distribution. The obtained results underscore the importance of detailed surface chemistry modeling, boundary layer resolution, and multi-dimensional flow effects in the accurate prediction of plasma–surface interactions. The simulation framework developed in this work offers a robust foundation for future experimental validation, coupling with material response models, and the design of advanced TPS configurations for reentry applications
Green ties, media eyes, and financial tides: exploring the dynamics of corporate environmental sustainability through networks, innovation spillover, and media influence
No full textSubmission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2027-08-01The student, Nima Safaei, accepted the attached license on 2025-07-17 at 04:51.The student, Nima Safaei, submitted this Dissertation for approval on 2025-07-17 at 04:53.This Dissertation was approved for publication on 2025-07-18 at 13:28.DSpace SAF Submission Ingestion Package generated from Vireo submission #22624 on 2025-10-25 at 15:54:16This dissertation examines how external actors shape firms’ environmental strategies through mechanisms of media scrutiny and supply chain relationships. Drawing on institutional theory, the relational view, and resource-based perspectives, it employs large-scale text analytics, structural econometrics, and graph-based deep learning frameworks to study three related topics: In the first study, “No News about Climate Action Is Good News for Low-Polluting Firms,” I examine how media coverage of firms’ climate actions impacts their operational and financial performance. Using a machine learning-based approach to capture the intensity of media attention on climate actions, the study reveals that increased media scrutiny can lead to higher operational costs and adverse financial performance for low-polluting firms. These findings highlight the need for corporate leaders to strategically manage media engagement around their climate initiatives. In the second study, “The Green Ripple Effect: Unraveling Green Innovation Spillover in Supply Chain Networks,” I examine the spillover of green innovation across supply chains, highlighting how the intensity of environmental patents by customers and suppliers influences the focal firm. Using a structural econometric model on a comprehensive dataset of supply chain linkages, the study uncovers a significant peer influence channel through which the green innovation intensity of customers and suppliers directly impacts that of the focal firm. This spillover effect highlights the interconnected nature of firms in their pursuit of sustainability. In addition to unveiling the significant role of peer influence in the spillover of green innovations, this research also demonstrates the positive impact of a shift toward more green innovations on a firm’s financial performance, further elucidating the competitive advantage it brings. In the final study, “Predicting Firm Greenwashing through Dynamic Supply Chain Graphs and Linguistic Cues in Job Postings,” I propose a novel methodology for forecasting corporate greenwashing by leveraging structural features from firms’ supply chain networks and linguistic patterns in their green job postings. The study employs a Dynamic Graph Neural Network (GNN) framework to capture evolving inter-firm relationships and time-varying firm attributes, enabling more accurate prediction of greenwashing tendencies based on network structure and firm-level cues. Additionally, I analyze job posting language, finding that firms with higher predicted greenwashing propensities tend to use vaguer and less concrete terms in their sustainability-related postings. This integrated approach highlights how combining dynamic network modeling with natural language processing can enhance our ability to detect greenwashing, offering valuable tools for promoting corporate accountability and transparency. Collectively, these studies contribute to a critical understanding of the complex interplay between corporate strategies, environmental sustainability, and external influences. By integrating insights from relational view theory of the firms, resource-based views, and institutional pressures, this dissertation offers theoretical corroboration of several organizational management theories through data-driven approaches and provides practical guidance for businesses striving to align their operations with sustainability goals. The research highlights the importance of strategic network formation, effective media management, and innovation spillover as pivotal elements that can significantly impact a firm’s ability to successfully adopt sustainable practices and reap economic benefits
Modular tendon routing for continuum manipulators
No full textSubmission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2027-08-01The student, John Golden, accepted the attached license on 2025-07-21 at 13:32.The student, John Golden, submitted this Thesis for approval on 2025-07-21 at 13:37.This Thesis was approved for publication on 2025-07-22 at 10:14.DSpace SAF Submission Ingestion Package generated from Vireo submission #22680 on 2025-10-25 at 15:54:26Tendon-driven continuum manipulators (TDCMs) exhibit inherent compliance akin to biological structures such as an elephant trunk or a octopus tentacle. Despite these advantages, TDCMs commonly struggle in providing the necessary strength to lift heavy loads, and reach a large number of end-effector positions in multiple different orientations. New workspace and deformation profiles can be achieved through alternative tendon-routing configurations: parallel, helical, hybrid. However, such configurations require predefined routing prior to deployment and are strictly limited to their defined metrics. In this thesis, two novel TDCMs are presented featuring subsequent modular disks capable of rotation and locking prior to (a priori) or following manipulator deformations (automated reconfigurable). Rotating one, two or multiple disk angles leads to various non-continuous segmented deformation of the backbone, leading to regions of unique curvatures. Analysis of the curvature is done through both multiple camera reconstruction of the backbone, or through manually recording each position of the disk. A nth-dimensional polynomial is fit through these points and Frenet-Serret theorem is utilized to extract bending, and torsion, values for a dataset of values rotated to a maximum of \pm{}90\si{\degree}. This provides the insight of which modular disks provides the biggest change to curvature. Multiple applications on the a priori manipulator are shown, in which the manipulator achieves common routing paradigms, obstacle avoidance, and taking multiple pictures from different 3D quadrants. The automated reconfigurable manipulator showcases other possibilities such as alternating between specific configurations while held at a fixed tendon displacement. Through the increase of the actuation space through localized modular disk, a larger amount of unique curvatures can be achieved thus improving metrics
