sigma-Type cyclopropenium cations (CPCs), which are formally generated by removing one substituent on the alkene of cyclopropenes, represent promising intermediates for forging functionalized cyclopropenes. However, sigma-type CPCs are challenging to access and their synthetic utility remains largely unexplored. Recently, we introduced electrophilic cyclopropenyl-gold(III) species as equivalents of sigma-type CPCs, which can then react with terminal alkynes or vinylboronic acids to afford alkynyl- or alkenyl-cyclopropenes. The merging of redox gold catalysis and a new class of hypervalent iodine reagents-the cyclopropenyl benziodoxoles (CpBXs)-plays a central role in generating the sigma-type CPC equivalents. With the same reagents, we have also developed a synergistic Au/Ag bimetallic catalytic cyclopropenyl cross-coupling system that enables 1,1 '-bicyclopropenyl derivatives to be forged in an efficient and modular manner. Our strategy provides access to previously inaccessible, yet highly useful, functionalized cyclopropenes, thereby significantly advancing both cyclopropene and hypervalent iodine chemistry. 1 Introduction
2 Transferring sigma-Type CPCs to Terminal Alkynes 3 Transferring sigma-Type CPCs to Vinylboronic Acids 4 Transferring sigma-Type CPCs to Terminal Cyclopropenes 5 Reaction Mechanism 6 Conclusions

Cyclopropenes are the smallest unsaturated carbocycles. They possess exceptionally high ring strain, resulting in unique reactivity, enabling C═C bond functionalization and ring-opening transformations. Herein, we report a copper-mediated trifluoromethylation of cyclopropenyl benziodoxole (CpBX) reagents, providing access to previously inaccessible, yet highly useful, sp2-trifluoromethylated cyclopropenes. Subsequent functionalizations of these novel strained ring building blocks enable access to highly substituted trifluoromethylated cyclopropanes, difluoromethylene cyclopropanes and noncyclic trifluoromethylated compounds.

The 2019 Undergraduate Biology Education Research Gordon Research Conference (UBER GRC), titled “Achieving Widespread Improvement in Undergraduate Education,” brought together a diverse group of researchers and practitioners working to identify, promote, and understand widespread adoption of evidence-based teaching, learning, and success strategies in undergraduate biology. Graduate students and postdocs had the additional opportunity to present and discuss research during a Gordon Research Seminar (GRS) that preceded the GRC. This report provides a broad overview of the UBER GRC and GRS and highlights major themes that cut across invited talks, poster presentations, and informal discussions. Such themes include the importance of working in teams at multiple levels to achieve instructional improvement, the potential to use big data and analytics to in-form instructional change, the need to customize change initiatives, and the importance of psychosocial supports in improving undergraduate student well-being and academic success. The report also discusses the future of the UBER GRC as an established meeting and describes aspects of the conference that make it unique, both in terms of facilitating dissemination of research and providing a welcoming environment for conferees.

Disciplinary silos and large amounts of specialized information in chemistry and biology courses undermines how students can make sense of these disciplines. This manuscript reports on how mechanistic reasoning across undergraduate courses may guide students towards more enduring and meaningful science learning. This team of chemistry, biology, and education researchers engaged in conversation about core mechanisms important for student learning in each disciplinary area that would connect to core mechanisms in the other disciplinary areas. The team also engaged in design work around written mechanistic explanations assessment items in each area. Those discussions prompted awareness of disciplinary and pedagogical similarities and differences about mechanisms. Our findings report on the mechanistic reasoning we focused on in each disciplinary area and how those were embodied in the assessment prompts. We discuss implications for teaching students who are traversing different subject matter information and ways of knowing.

Causal mechanistic reasoning is a thinking strategy that can help students explain complex phenomena using core ideas commonly emphasized in separate undergraduate courses, as it requires students to identify underlying entities, unpack their relevant properties and interactions, and link them to construct mechanistic explanations. As a crossdisciplinary group of biologists, chemists, and teacher educators, we designed a scaffolded set of tasks that require content knowledge from biology and chemistry to construct nested hierarchical mechanistic explanations that span three scales (molecular, macromolecular, and cellular). We examined student explanations across seven introductory and upper-level biology and chemistry courses to determine how the construction of mechanistic explanations varied across courses and the relationship between the construction of mechanistic explanations at different scales. We found non-, partial, and complete mechanistic explanations in all courses and at each scale. Complete mechanistic explanation construction was lowest in introductory chemistry, about the same across biology and organic chemistry, and highest in biochemistry. Across tasks, the construction of a mechanistic explanation at a smaller scale was associated with constructing a mechanistic explanation for larger scales; however, the use of molecular scale disciplinary resources was only associated with complete mechanistic explanations at the macromolecular, not cellular scale.

Aerosols influence Earth's climate through direct interactions with radiation and by modifying cloud properties via their effects on cloud droplet and ice crystal formation. Uncertainties associated with aerosol-radiation and aerosol-cloud interactions represent the largest physical uncertainty in quantifying climate change, arising from the complexity of aerosol processes, limited observations, poor constraints on pre-industrial aerosol, and incomplete representation of aerosol processes in climate models. These uncertainties are particularly relevant in the Arctic, where climate change is amplified and aerosol-cloud and aerosol-radiation interactions play a key role in the surface energy balance. This thesis focuses on two natural aerosol sources in the Arctic, sea salt aerosol associated with blowing snow and mineral dust from high-latitude glacial outwash plains, with the aim of characterizing their climate-relevant properties. Both aerosol sources can exert strong climate impacts as ice-nucleating particles (INPs), cloud condensation nuclei (CCN), or contributors to scattering, yet remain poorly constrained by Arctic observations. Comprehensive observations from the year-long Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) expedition (2019-2020) were used to examine wind-driven aerosol production in the central Arctic. Periods of high wind speed and blowing snow were associated with strong enhancements in aerosol number concentrations, submicron sodium chloride mass, CCN concentrations, and scattering coefficients, indicating the potential climate relevance of this process. Seasonal differences and the influence of snow age on aerosol enhancements highlight the dependence on season and surface conditions. Accounting for air mass history demonstrates the role of regional transport in explaining coarse-mode aerosol variability. Together, these results provide direct observational evidence for wind-driven aerosol production in the central Arctic and offer constraints for model evaluation. Glacier retreat across the Arctic exposes new land surfaces and leads to the expansion of glacial outwash plains rich in fine glacial dust, which can influence cloud properties upon emission. To investigate high-latitude dust as an INP source, a two-month field campaign was conducted in southwestern Greenland in 2023. To support these measurements, the Sion Particle Ice Crystallization Experiment (SPICE), a droplet freezing assay for measuring immersion-mode INPs, was developed and validated in the laboratory. Glacial dust in southern Greenland showed ice nucleating activity relevant for mixed-phase clouds, but with lower activity compared to other high-latitude dust sites, highlighting variability in the Arctic. Small amounts of organic material were identified as the main driver of ice-nucleating activity and its variability in the glacial outwash plains. Atmospheric INP measurements indicated a substantial biogenic contribution to the INP population and a local influence of the glacial outwash plain during summer.
Together, these results provide new observational constraints and process understanding of natural aerosol sources in the Arctic, highlighting their potential climate relevance and supporting improved representation of these processes in climate models.

Interactions of multiple agents lie at the heart of many real-world systems, including transportation networks, auctions, electricity markets, and multi-robot systems. Game theory provides a natural framework for modeling such strategic interactions. Historically, research in this field has focused on the existence and computation of equilibria, as well as on learning dynamics that allow agents to converge to stable outcomes. However, when applying such game-theoretic tools to real-world systems, several challenges emerge: In many settings, agents face constraints that render certain decisions infeasible. This is relevant in domains where agents are limited by shared resources, safety and fairness requirements, or physical constraints. A second challenge concerns system-level efficiency, as strategic behavior frequently leads to socially inefficient outcomes, such as congestion or resource overuse. In this thesis, we aim to broaden the applicability of the game-theoretic framework by addressing these challenges through novel equilibrium concepts and decentralized learning dynamics.

The first part of the thesis studies multi-agent systems subject to constraints. In Chapter 3, we consider coupling constraints, where an agent's feasibility depends on the joint strategy profile. We formalize the notion of a constrained correlated equilibrium and establish sufficient conditions for its existence in Markov games. In Chapter 4, we address the problem of learning in contextual games with private constraints, where feasibility is determined by an agent's own decision set. We propose a learning algorithm and derive sublinear convergence bounds for both cumulative regret and constraint violations. The effectiveness of this approach is demonstrated through a multi-building temperature control case study.

The second part of the thesis focuses on efficiency in multi-agent systems with welfare objectives. In Chapter 5, we study how agents can be steered toward socially desirable outcomes within a Stackelberg framework without prior knowledge of the agents' reward functions. We develop a no-regret learning algorithm for the leader and prove convergence to a Stackelberg equilibrium when followers play approximate Nash equilibria. This approach is evaluated in an electric ride-hailing pricing study. Finally, in Chapter 6, we consider the problem of equilibrium selection in potential games. While log-linear learning is known to converge asymptotically to efficient Nash equilibria, we provide the first finite-time convergence guarantees for these dynamics and establish robustness under reduced feedback and utility perturbations.

Continuous development of thin film deposition technologies is essential for fabricating materials that meet the targeted requirements of specific applications. In this context, the present dissertation focuses on atomic layer deposition (ALD) of metallic copper, aiming to deepen the understanding of Cu ALD and establish a groundwork for future application-driven process development. Specifically, the Cu ALD process investigated in this work was based on copper(II) hexafluoroacetylacetonate (Cu(hfac)2) and the metal-containing reducing agent diethylzinc (DEZ). To date, despite extensive efforts, the development of reliable metal ALD processes remains challenging due to several limitations, the most critical of which are: long nucleation delays, islands formation, and incomplete understanding of the surface reaction mechanisms. To address these issues, the reaction mechanism of Cu(hfac)2 and DEZ during ALD was first examined using in situ time-of-flight mass spectrometry (TOFMS). The investigation focused on the qualitative analysis of the surface reactions occurring during steady-state growth. The high sensitivity and mass resolution of TOFMS enabled the identification of previously only hypothesized surface reaction volatile by-products, revealing two different mechanisms in each ALD half-cycle. During the Cu(hfac)2 pulse, both Zn(hfac)2 and EtZn(hfac) volatile by-products were detected in varying amounts, while the DEZ pulse predominantly yielded EtZn(hfac). The majority of the reaction occurred during the DEZ half-cycle, suggesting the presence of an abbreviated ALD cycle. Building upon these mechanistic findings, copper deposition was successfully extended to various substrates, specifically on oxide, nitride, and flexible materials. The resulting Cu films exhibited compact but island-like morphology, consistent with a Volmer-Weber growth mechanism assisted by surface diffusion of copper species. Interestingly, no growth was observed on metallic surfaces such as Au and Pt, highlighting the inherent selectivity of this specific Cu ALD process. This selective behaviour was further explored using patterned metal-oxide and metal-nitride substrates, demonstrating inherent selective ALD (AS-ALD) on micro-scale device geometries. Cu deposition was also attempted on high-aspect-ratio (HAR) structures, demonstrating initial conformality, but requiring further development to achieve uniform coverage. TOFMS investigation was further extended by monitoring 400 Cu ALD cycles to provide insights into possible changes in the reaction pathway and by products yields from the early stage to the steady state regime. Initial experiments focused on oxide substrates, while future work will extend these studies to other surfaces to assess the substrate influence. In conclusion, by linking mechanistic insights with process optimization, this work enabled controlled and reproducible Cu ALD on diverse substrates. The preliminary Cu depositions on technologically relevant substrates, such as flexible materials, gas diffusion layers for catalysis, and dental implant materials, highlight the practical relevance of this study. Overall, these findings lay the groundwork for future application-driven Cu ALD processes.

The urgent need to decarbonize Europe's energy sector has placed electrification and the deployment of renewable electricity production at the forefront of climate change mitigation strategies. The integration of intermittent and distributed renewable energy sources, such as solar photovoltaic (PV) electricity, into the power grid will require both increased flexibility in electricity demand and adequate storage capacity. On the other side, the electrification of private passenger mobility will further increase the need for decarbonised electricity. At the same time, electric vehicles (EVs) could represent an opportunity to enhance flexibility in electricity consumption and serve as local storage for renewable electricity production. By shifting EV charging during the day, it can contribute to absorbing PV production peaks and mitigate additional demand during peak periods. Furthermore, the EV batteries can be discharged to the grid or home during the evening or other demand peaks via vehicle-to-grid (V2G) strategies. However, the use of EVs, PV and V2G requires deep changes in our approach to energy systems and personal mobility. For this reason, it is of primary importance to make models available highlight the potential of V2G in reducing CO2 emissions, lowering costs and increase our energy sovereignty. Moreover, policymakers and energy planners, especially at the local scale, need to have access to tools to help them plan the electrification of the society and projection of the effects of their decisions.

This thesis addresses the challenge of modelling the large-scale electrification of private mobility in Europe until 2050, with a focus on charging needs, charging infrastructure and the synergies between EVs and PV. The approach consists in the geographic and temporal modelling of mobility and charging needs, that are then compared temporally with a modelled PV production, to quantify the synergies of EV and PV considering different charging strategies (smart, V2G) while estimating the associated required charging infrastructure. The work presentes a comprehensive methodology, based only on open data, to estimate the mobility needs, charging needs, PV coupling potential and charging infrastructure deployment in Europe up to 2035. The models developed in this work are implemented in the open source online platform citiwatts.eu, which can be used by anyone to create user-defined scenarios and projections. Thus, this work provides policymakers, energy planners, and other stakeholders with insights to design sustainable mobility systems.

Metasurfaces are nanophotonic elements that allow for the controlled manipulation of light. Many applications require control of both the angular and spectral responses of a single metasurface, for example, in light trapping for thin-film photovoltaic absorbers. Existing design approaches often rely on iterative optimization and do not provide interpretable handles at the design stage for controlling angle and spectral responses.

This thesis introduces and validates a design framework for correlated-disordered metasurfaces that provides design-stage control over angular and spectral scattering responses. The underlying correlated-disordered point lattice is obtained by prescribing spatial correlations directly in Fourier space and discretizing the real space into a set of coordinates. The prescribed spatial correlations are preserved in the discretization and determine the angular scattering response of the ensemble. The lattice is decorated with nanoscatterers, whose geometries determine each nanoscatterer's resonant response and thereby modulate the metasurface's spectral response. Angular scattering control is experimentally validated using Fourier scatterometry, while spectral behavior is computationally assessed.

The framework is extended to the interleaving of two independently designed correlated-disordered metasurfaces into a single plane. The combined angular scattering response is approximately additive, as shown experimentally. Computational results indicate a similar preservation of the spectral scattering response, revealing that the individual scattering behavior is largely preserved during the interleaving step, with observable cross-talk between the two metasurfaces. A systematic computational study further investigates how variations in the key design parameters within this framework influence the collective scattering response.

Light trapping in thin-film photovoltaic absorbers is an established problem in nanophotonics, where angular and spectral control of incident light can improve performance by coupling it to guided-mode resonances. The design framework enables the realization of multi-resonant, correlated-disordered light-trapping patterns by interleaving two metasurfaces with distinct spatial correlations and resonator geometries, thereby combining prescribed angular redistribution with discrete multi-resonant elements. Compared with double-periodic patterns with the same two resonator geometries and with interconnected correlated-disordered patterns that share similar structural correlations but lack discrete resonators, this approach yields higher absorption in a thin-film absorber, as revealed by computational investigation.

Zinc phosphide is chosen as a demonstrator material in this thesis. The stability of this emerging photovoltaic material, made from earth-abundant elements, under atmospheric conditions is investigated. A long-term experimental study identifies surface oxidation as the dominant degradation mechanism, which can be mitigated by a thin dielectric capping layer.

Together, these results establish a non-iterative and interpretable design methodology for correlated-disordered metasurfaces by combining the reciprocal-space prescription with discrete resonator decoration and interleaving strategies. Demonstrated for thin-film light trapping, the approach generalizes to metasurface design problems that benefit from the joint engineering of spatial correlations and resonant building blocks.