50107 research outputs found
Sort by
RADICAL POLYMERIZATIONS IN WATER: EMULSION AND SEMI-BATCH PROCESSES TO YIELD FUNCTIONAL POLYMER PARTICLES AND INTERFACE MODIFIERS
No full textPerforming free radical polymerization in aqueous environments is of great interest in fundamental polymer science as well as scalability at the industry level. This thesis focuses on free radical polymerization in water, using techniques of emulsion polymerization and semi-batch polymerization to prepare functional polymer particles and interface-active agents.
In the Introductory chapter, Chapter 1, fundamental considerations of free radical polymerization in aqueous media are described. This is followed by the key role of free radical polymerization in heterogeneous polymerization methods, such as emulsion polymerization and the complexities associated with the preparation of functional polymer particles. The focus then shifts to fundamental aspects associated with employing free radical polymerization in “semi-batch” techniques. The introductory chapter is followed by two chapters comprising the core of my thesis research involving free radical polymerization and polymer particles containing neutral, charged or zwitterionic monomers. In Chapter 2, the synthesis of alkyne-rich, patchy polymer colloidal particles is described, using the key synthetic technique of surfactant free emulsion polymerization (SFEP). Polymer colloids synthesized by this technique were subjected to azide-alkyne cycloaddition reactions to alter the functionality on the particle surface and interior. In Chapter 3, the focus shifts to the preparation of hydrophilic polymers containing both charged and uncharged moieties by semi-batch free radical polymerization in water, with the resultant polymers anticipated to serve as useful interface-active agents.The work described in this thesis was made possible by support from the National Science Foundation (NSF-CHE-2203578) as well as DuPont.Doctor of Philosophy (Ph.D.)2026-09-0
The Forgotten Men: African and Colored Servicemen of the Second World War and the Rise of African Nationalism in Southern Rhodesia (Colonial Zimbabwe), c.1939-1980.
No full textThis dissertation examines the role played by the Second World War in rise of African nationalism. It specifically examines the mobilization, recruitment, wartime, demobilization, and post-war experiences of African and Colored servicemen who participated in the war from Southern Rhodesia (Colonial Zimbabwe). It examines how their experiences revived African nationalism from the 1940s to the late 1970s. It argues that the war engendered political awareness, anti-colonial sentiments, attitudes, and eventually political activity that aided the rise of African nationalism and the attainment of its goals in Southern Rhodesia. The bulk of the material for this dissertation was obtained from archival sources, written family records and oral interviews.Department of HistoryDoctor of Philosophy (Ph.D.)2030-09-0
POLYMER-BASED MULTICHANNEL SENSOR ARRAYS FOR ADVANCING CELL-BASED DIAGNOSTICS AND DRUG DISCOVERY
No full textMammalian cell surfaces are intricate and dynamic, composed of proteins, lipids, and glycans that change during disease progression and cellular responses. These molecular features offer rich diagnostic and therapeutic potential. Traditional “lock-and-key” sensors, which rely on highly specific interactions, often achieve excellent selectivity but suffer from limitations in flexibility, multiplexing capability, and applicability across diverse cell states.
Inspired by the human olfactory system, multichannel sensor arrays provide an alternative approach. These arrays use cross-reactive sensing elements that interact differentially with analytes, generating unique fluorescence fingerprints. When paired with machine learning algorithms, these patterns enable sensitive, hypothesis-free, and high-throughput classification of analytes. Synthetic polymers are particularly attractive for building such arrays due to their chemical stability and tunable properties. With supramolecular chemistry principles, polymers can be designed to produce diverse and responsive sensor platforms capable of discriminating subtle variations in biological environments.
In this dissertation, I present the development of cross-reactive polymer-based sensor arrays for cellular diagnostics and drug discovery. Initial work involved polymer-based sensor arrays that distinguished diverse mammalian cell types and states. A Förster Resonance Energy Transfer (FRET)-based array enabled sensitive detection of macrophage responses to femtomolar pesticide levels. We further utilized specific supramolecular interactions between cell surface glycan diols and boronic acid groups to construct a pH-responsive poly(oxanorborneneimide)-based sensor array. This system discriminated glycosylation phenotypes and identified drugs capable of repolarizing macrophages toward anti-inflammatory states. Building on these findings, a novel one-polymer multichannel nanosensor was developed for rapid and accurate identification of varying ratios of cancer-associated fibroblasts (CAFs) within fibroblast populations and tumor-associated macrophages (TAMs) within macrophage populations, two crucial but often challenging targets in cancer diagnosis and tumor treatment. The final section presents the development of a multiplexed imaging strategy by integrating an array-based sensor with imaging flow cytometry for precise cell identification and detection of low-level contamination.
Taken together, these studies highlight the potential of array-based polymeric sensing platforms as a robust and scalable platform for advancing cellular-level risk assessment, diagnostics, and therapeutic discovery. This strategy opens new avenues for unraveling complex biological systems and developing innovative tools with broad applications in biomedical research and clinical practice.Doctor of Philosophy (Ph.D.)2026-09-0
Uncovering the Roles of Charge Extraction and Passivation at Interfaces in Perovskite Solar Cells
No full textPerovskite solar cells (PSCs) have emerged as a leading thin‑film photovoltaic technology, delivering power conversion efficiencies (PCE) exceeding 27% with the promise of low‑cost, solution‑based processing. However, their long‑term operational stability under illumination is undermined by mobile ions and interfacial defects. The impact of the critical interfaces between the perovskite layer and the adjacent charge-selective layer on light-induced instability remains insufficiently understood. In this dissertation, we have set out to systematically investigate the impact of chemical and electronic properties of charge-selective materials on the ion migration, charge extraction, and light‑induced degradation in PSCs. We introduce a tailored hole‑transport bilayer and develop a stability‑monitoring platform that together dissect the complex interplay among carrier extraction dynamics, defect passivation, illumination direction, and ion transport. By engineering interfacial chemistry, we suppress both anion and cation migration, achieving a champion PCE of 24.4% and demonstrating excellent operational stability under continuous illumination. Our findings reveal the mechanistic origins of light‑induced degradation and establish clear design guidelines for interfacial engineering, paving the way toward durable, high‑efficiency perovskite photovoltaics suitable for commercial deployment.US National Science Foundation (NSF) DMR 2101127Doctor of Philosophy (Ph.D.
(De)Centering Whiteness: A Betweener Conversation
No full textUntil March 2020, I moved easily and seamlessly between my two roles and contexts that couldn’t be more different: a doctoral student at UMass, Amherst in Amherst, Massachusetts; Executive Director of Jitegemee Children’s Program (Jitegemee), a community development program in Machakos, Kenya. And then the COVID-19 pandemic struck. Flights stopped; work that could, shifted online; we were limited to our homes; travel was out of the question. During the time we were isolated and forced to stay indoors, there was the opportunity for reflection and considering new ways of being. Following the police murder of George Floyd and the summer of 2020 BLM protests, the international development sector had its own overdue moment of reckoning, which led to online conversations, Zoom workshops, media reports and so on and a flurry of anti-racist statements from academic and NGO organizations alike.
In this dissertation I use performance autoethnography to offer my embodied critique of an ongoing process of (de)centering Whiteness in my academic scholarship as well as in my work as a development practitioner. That summer, I too was involved in my own personal moment of reckoning as a White development scholar-practitioner, born in apartheid South Africa, working between Kenya and the USA. How could I effectively lead an organization halfway across the world, situated in a completely different context, much less during a global pandemic that rendered visits also impossible? We started considering how we could realize a transition at Jitegemee to ‘local leadership,’ by then a tired sector buzzword that was rarely implemented, in ways that would minimize disruption and harm to the organization and children we supported. This dissertation traces that back-and-forth journey, in community, in conversation, engaged in a process of care and collaboration. I share this dissertation as an offering of love and solidarity; an offering about a process that is incomplete and ongoing. These reflections, our experiences, and challenges with this work, are a snapshot of what I/we did, where I/we were, and how I/we individually and collectively wrestled with the complexities of the on-the-ground, lived realities of what it meant to us to (de)center Whiteness. You are invited to join this journey.Doctor of Philosophy (Ph.D.)2026-09-0
Human-Centered Design of Contextually Appropriate Conversational Agents for Enhancing Public Interaction, Engagement, and Multimodal Experiences
No full textDesigning conversational agents that can engage users, adapt to diverse contexts, and communicate complex information effectively is a growing focus across domains such as healthcare, digital civics, and education. AI-enabled conversational agents have emerged as a promising alternative to traditional methods due to their natural, engaging, and personalized interactions that can elicit richer, more meaningful feedback. Despite these advancements, three important challenges remain for the human-centered design of conversational agents. First, in conversations involving multiple topics, existing systems struggle to manage the conversation flow and context across the topics, resulting in interruptions and lower-quality responses. Second, the current literature lacks investigation into the design of conversational agent avatars, and we currently lack guidelines to inform avatar design and its impact on user trust, usability, and engagement. Third, although personalization and multimodal communication can boost a system’s relevance and effectiveness, we still know little about how to integrate them in ways that maintain accuracy, scale efficiently, and fit the context appropriately. This dissertation addresses these challenges through three connected investigations: designing mechanisms for seamless multi-topic conversation management, mapping the design space of avatar features, and developing agents that deliver dynamic, personalized, and context-aware interactions.
The first part of this dissertation focuses on keeping conversations flowing smoothly across multiple topics. In many real-world applications, users switch between related but distinct topics such as personal health, community issues, and environmental concerns within a single session. In these settings, single-agent chatbots often struggle to accurately interpret complex, multi-faceted user input and capture nuanced information across diverse topics. To address this, I investigated the use of multi-agent chatbots for multifaceted civic data collection, which reduces computational cost by distributing topics among multiple agents. To address this, I designed and developed CommunityBots, a multi-agent conversation platform where each agent specializes in a specific conversation topic. To manage conversation across multiple topics and agents, I introduced a novel Conversation and Topic Management (CTM) mechanism that handles topic-switching and chatbot-switching based on user responses and intentions. A between-subjects study with 96 crowd workers demonstrated that CommunityBots significantly increased user engagement, improved response quality, and reduced conversational interruptions compared to a single-agent baseline. The visual cues integrated into the interface helped participants better understand the functionalities of the CTM mechanism, resulting in increased user satisfaction.
Building on these findings, the second part of this dissertation investigates how the agent's visual representation influences interaction. While avatars are frequently used to boost engagement, there is little guideline about how specific avatar features influence user perceptions across domains, devices, and conversation styles. To gain a more comprehensive understanding of the conversational agent avatar design space, I conducted a thorough analysis of existing literature to map the avatar design space. I defined a categorization of 10 dimensions that is based on the analysis and iterative coding of 266 conversational agent papers from 160 venues spanning 2003 to the present. In addition, I built an interactive browser to facilitate exploration and interaction with these dimensions and their interrelationships. This categorization lays the groundwork for researchers, designers, and practitioners to discern task-specific and contextual aspects of conversational agent design.
The final part of the dissertation explores how personalization and multimodality can make conversational agents more dynamic and context-aware. I first applied this approach in PATRIKA, an AI-enabled conversational journaling system for people with Parkinson’s disease, which integrates cooperative conversation principles, clinical interview simulations, and personalization to capture clinically valuable information. Two user studies showed that PATRIKA enhanced engagement and supported more complete, accurate journaling compared to traditional methods. Building on the lessons from designing personalized, context-sensitive prompts in a healthcare setting, I extended these principles to the domain of climate change communication through CLAImate. This system delivers personalized and localized climate change narratives and visualizations, adapting to a user’s location, prior knowledge, and interests. Internal testing, a formative study, and a user study demonstrated that this approach made climate information clearer, more relevant, and easier to understand.
These contributions advance conversational AI by demonstrating how multi-agent designs, avatar design, and personalized, bidirectional interactions can boost engagement and data quality in high-impact domains. The proposed frameworks and systems provide practical approaches to developing more adaptive, context-aware agents. This work also lays the groundwork for future research into scaling such systems for real-world applications.My research on managing multi-topic conversational flow through CommunityBots was supported by the National Science Foundation (Grant CNS-2125183), with additional funding from the Center for Data Science and the ADVANCE Program at the University of Massachusetts Amherst.Doctor of Philosophy (Ph.D.
Age Dependent Prefrontal Dynamics of Reappraisal Yield Comparable Cognitive Outcomes
No full textEmotion regulation shapes not only how people experience emotional events, but also how those events are remembered, with lasting consequences for behavior and well-being. Reappraisal - the reinterpretation of negative stimuli as more positive – boosts encoding of negative episodes due to a deeper engagement with the meaning of emotional content, while attenuating negative affect over time. Among young adults, reappraisal recruits a distinct network of limbic and cognitive control regions that enhance memory for emotional episodes, thus guiding adaptive resolution of similar negative scenarios in the future. Aging is marked by a greater motivation to regulate emotion and preserve well-being, relative to younger adulthood. Yet, little research has examined if reappraisal reduces negative emotion and enhances memory for negative scenarios in ways that may be similar to -or distinct from- those behavioral and neural mechanisms observed in younger adults. Thus, this dissertation addresses a theoretical gap in clarifying the neural substrates of encoding during reappraisal that support long-term emotional memory in both young and older adults. Twenty young (18-29 years) and 19 older (60-75 years) adults underwent incidental encoding of negative and neutral images during an event-related fMRI scan. Participants either reappraised negative images to reduce negative affect or viewed negative and neutral images passively. Recognition memory for images was assessed in the scanner after a 10-minute (i.e., immediate recall), and 4-hour delay (i.e., delayed recall). Recognition of images did not significantly vary by age group or regulatory condition. However, older adults demonstrated higher recognition sensitivity to neutral relative to negative images. Young adults demonstrated greater relative recruitment of distributed frontal/prefrontal, temporoparietal, and occipital regions to support reappraisal, whereas older adults comparatively relied on a small subset of parietal and posterior cortical regions supporting attentional control and visual integration. In an age-invariant manner, greater prefrontal recruitment predicted worse subsequent recognition accuracy, suggesting a tradeoff between top-down cognitive control and mnemonic efficiency during passive encoding of emotional information. Taken together, these findings reveal neural signatures of reappraisal among both age groups that may interfere with incidental memory encoding for emotional experience and regulation.Doctor of Philosophy (Ph.D.)2026-03-0
Investigation of potential and known weedy rice (Oryza sativa) competitiveness traits: beneficial plant-microbe relationships and the genetic basis of seed shattering
No full textCultivated rice is a staple food crop grown worldwide. A major threat to the production and yield of cultivated rice is due to a closely related agricultural weed, weedy rice. Most weedy rice populations have evolved through the process of de-domestication of different cultivated rice varieties grown worldwide. During this process, these populations have gained competitive traits such as increased seed shattering ability. In this dissertation thesis, I sought to determine if there are potential physiological differences and genetic mechanisms responsible for the increased competitive ability of weedy rice in cultivated rice fields. In Chapter 1, I determined and compared the root endophytic microbial community composition, diversity, and richness of multiple genotypes of weedy and cultivated rice. I found no distinct differences between community composition, however, cultivated rice was found to have higher bacterial richness and diversity. In Chapter 2, I isolated the culturable seed endophytic microbiome of weedy and cultivated rice and assessed the isolates for four plant growth-promoting abilities: nitrogen fixation ability, indole-3-acetic acid production, phosphate solubilization, and iron scavenging ability. We found microbial isolates from these weedy or cultivated rice groups do not have distinctly different plant growth-promotion ability. The top performing microbial isolates were used as seed bioinoculants and we demonstrated a differential response in seedling growth between weedy and cultivated rice. In Chapter 3, I conducted a quantitative trait locus mapping study to identify potential genomic regions that are responsible for seed shattering in weedy rice. In this work, I identified a marker that was identified as a single nucleotide polymorphism previously suggested as influential on seed shattering in weedy rice. We found that this G to T mutation was correlated with a high shattering ability and a defined abscission zone in our mapping population and in created hybrid inbred families. Overall, these findings build upon our foundation of weedy rice competitive ability which is important for the creation of novel mitigation strategies.Doctor of Philosophy (Ph.D.
THE EFFECTS OF SEA LEVEL RISE ON THE HABITAT OF COASTAL WATERBIRDS IN NORTHWEST MEXICO.
No full textIn my introductory chapter, I reviewed how coastal biodiversity may respond to sea level rise (SLR), including potential expansions, shifts, contractions, or extirpations. Using a trait-based framework, I identified characteristics that may make species more or less vulnerable to SLR. Coastal waterbirds emerged as key indicators for future research due to their high vulnerability to habitat loss.
Chapter 1 presents a regional assessment of SLR impacts in Northwest Mexico—one of the most important areas for waterbirds along the Pacific Flyway. Using the red knot (Calidris canutus roselaari) as a model species, I found that 55% of current wetland habitat could be permanently submerged by 2050, while remaining high-quality habitat may lose 20% of its suitability. These changes could accelerate red knot declines and affect many other species.
Chapters 2, 3, and 4 explore how sea level rise affects all aspects of the waterbird life cycle—breeding, feeding, and roosting—revealing vulnerabilities across each essential stage.
Chapter 2 focuses on the breeding impacts of SLR for two threatened beach-nesting birds in Ensenada, Mexico: the Least Tern (Sternula antillarum) and Snowy Plover (Anarhynchus nivosus). Tidal flooding already causes 18% of nest failures. However, nests on berms—elevated sand platforms—had 76% higher survival, suggesting berms are a key feature to mitigate flood risk.
In Chapter 3, I evaluated how long-term SLR could impact foraging habitats in Guerrero Negro Lagoon for a 28-species waterbird community. By 2100, red knots may lose 43% of their low-tide foraging habitat and experience 86% less foraging time. Thirteen small-bodied species are projected to be affected a decade earlier than larger species, indicating cascading community-wide effects.
Chapter 4 documents a novel strategy used by waterbirds during spring tides: ephemeral roosts on floating rafts of dead seagrass. These small rafts supported up to 60 birds/m² and 25 species, representing key roost habitats during high tides—a proxy for SLR. Their effectiveness suggests potential for nature-based solutions to mitigate roosting habitat loss.
Overall, my dissertation underscores the urgent need to plan for both immediate and long-term impacts of SLR, with species-specific insights and nature-based solutions critical for conserving vulnerable coastal bird populations.Doctor of Philosophy (Ph.D.
Urban Air Pollution and Climate Stressors: Exploring Nitrogen Dioxide Patterns and Heat Wave Impacts in Holyoke, MA
No full textClimate change and air pollution are interconnected, with nitrogen dioxide (NO2) and heatwaves intensifying in urban environments, exacerbating public health risks and impacting urban agriculture. Therefore, we aimed to conduct a city-wide year-long measurement campaign to understand the spatiotemporal patterns and concentrations of NO2. Also, we aimed to assess the impact of heatwave exposure during early plant development on plant growth, phenology, flowering, fruit production, and the abundance of floral and non-floral visitors. We measured ambient NO₂ concentrations in Holyoke, MA, from August 2021 to 2022, using Ogawa passive samplers. We collected 260 weekly NO2 samples at 25 sites classified based on their land use characteristics. Sites included ‘fixed’ (yearlong, n = 5) and ‘rotating’ (weeklong, n = 20) locations. Results showed higher NO2 concentrations during winter (>150 µg/m³) compared to summer (<3.4 µg/m³), with commercial and highly dense sites experiencing the highest mean concentrations (17.09 µg/m³). Approximately 61% of rotating sites and 62% of weekly samples from fixed sites exceeded WHO air quality guidelines. A mixed-effects model revealed that weekly mean temperature, NDVI, and wind speed were the strongest predictors of NO₂. Additionally, we investigated the effects of heatwaves in urban environments using cherry tomato plants as a model system. Plants (n = 96) were exposed to heatwave (HW:37.7°C) and control (C: 28.8°C) conditions for 3 days during early development, then transplanted to pots and moved to 3 sites. We performed various surveys to assess plant growth, flowering, phenology, fruit production, and floral and non-floral visitors’ abundance. Heatwave exposure increased total flower counts yet did not influence initial growth rate or weekly plant height, but when the site with additional nutrient access was excluded, HW exposure significantly reduced plant height. HW-exposed plants produced more mature and immature fruits, attracted more non-floral visitors, and led to a delay in fruit ripening time. These findings highlight the dual challenges of air pollution and climate stressors in urban environments, emphasizing the need to understand their interactions to protect public health and support urban agriculture.JPB Foundation through Harvard’s Environmental Health ProgramMaster of Science (M.S.