Environmental and Occupational Health Sciences Institute
Rutgers University Community RepositoryNot a member yet
58345 research outputs found
Sort by
Scanning tunneling microscopy studies of twisted van der waals layers
No full textEver since the discovery of atomically thin two-dimensional crystals, techniques have beendeveloped to manipulate their properties, among which constructing Van der Waals
heterostructures by twisting and stacking identical or different monolayers has gained extensive
attention. The simplest case of stacking two atomic crystals with a twist between their crystal
axes produces moiré potentials leading to superstructures that can dramatically modify the
electronic properties. In this thesis I show that superposing three layers with a twist between their
crystal axes generates a rich variety of intricate structures and electronic properties.
By using scanning tunneling microscopy and spectroscopy to study crystals of twisted bilayer
graphene on hexagonal boron nitride (hBN), we unveil a super-moiré phase diagram spanned by
the lattice constants of the two superposed moiré lattices (graphene-on-graphene and grapheneon-
hBN), which contains both commensurate periodic and incommensurate quasiperiodic
crystals. 1:1 commensurate crystals, which should theoretically exist at only one point on this phase diagram, are observed over a range deviating by as much as 20% from the predicted value,
demonstrating an unexpected self-alignment mechanism. The incommensurate crystals include
quasicrystals, which are quasiperiodic and feature a Bravais-forbidden dodecagonal symmetry,
and intercrystals, which are also quasiperiodic but lack forbidden symmetries. This rich variety
of tunable double moiré structures offers a synthetic platform for exploring the unique electronic
properties of quasiperiodic crystals, which are rarely found in nature.Ph.D.Includes bibliographical reference
Winter warmth: quantifying historical changes in very warm winter days across the United States
No full textTemperatures have increased across the United States, and are projectedto continue to do so as a consequence of growing greenhouse gas
concentrations in the atmosphere. As temperatures have increased, the
temperature distribution has shifted, making extremely warm days more likely
to occur. Warm extremes are important because of their disproportionate
impacts on society. While warm extremes have been well studied in the
summer when heat is most deadly there is little in the literature about warm
extremes in the cold months, even though anomalous winter warmth can also
lead to significant environmental, economic, and cultural harm. To quantify
risks associated with extreme winter warmth in the United States, a better
understanding of the characteristics of these events is necessary. Using a
combination of surface observations from 200+ locations across the country, as
well as gridded surface and reanalysis data, historical variations and trends in
warm winter events since 1951 are examined. Very warm winter days are
defined as those with daily maximum temperature above the 95th percentile
threshold based on the temperature distribution for the early part of the record,
1951-1990. Overall, very warm winter days show widespread increases over
time across the United States, although, there are regional differences in the
rate of change of very warm winter days. These regional differences are largest
where increases in average winter maximum temperature are also largest with
the greatest increases taking place in the Northeast and Upper Midwest
regions. There is also evidence in some regions that the warmest winter
temperatures are changing at a faster rate than the median temperature
change, suggesting that the shape of the temperature distribution may be
changing. Similar patterns of trends in the frequency of very warm winter days
are generally present across all three datasets analyzed. However, one notable
difference among datasets is that on average the gridded surface data from
NOAA nClimGrid-Daily indicate an increase in very warm winter days at a
higher rate than other datasets. This is surprising given that NOAA nClimGrid-
Daily is based on the same observations that make up the station dataset. Given
that this thesis was a collaboration with the Climate Matters program at
Climate Central, the results are designed to be disseminated through the
nationwide network of meteorologists and journalists participating in Climate
Matters by including localized findings with nationwide context.M.S.Includes bibliographical reference
When less is more: the formal features of status downplay
No full textWhy do affluent people drive used cars, elite chefs praise basic ingredients, and upper-middle-class couples arrange backyard weddings? What makes senior professors feel comfortable dressing down and allows men to bring their kids into the office? This dissertation takes a formal sociological approach to explore the cultural phenomenon of “Status Downplay”—a paradoxical type of status display, where more social status is signaled through a subtraction of status symbols. Based on two-site research (United States and Israel), and using the innovative method of “trigger interviews” along with textual and discourse analysis of texts, images, and artifacts, I conduct a cross-cultural, trans-contextual, multi-perspective analysis of diverse manifestations of status downplay. Articulating the semiotic code of this abstained, indirect, and symbolic status signification, the social pattern analysis uncovers the formal features of status downplay: the implied semiotic contrast to “status-overdoing,” the cultural expectations of (additive) “doing” and default assumptions about (the presenter’s status-) “being,” and the semiotic reliance on cultural classifications of social markedness and unmarkedness—and relevance and irrelevance—that underly this parodical presentation of self. On this basis, I examine status downplay’s key semiotic code of performative affordance and asymmetrical semiotic structure—under which socially dominant and culturally unmarked actors are entitled to the social privilege of semiotic flexibility.Ph.D.Includes bibliographical reference
Peeking beneath Titan’s haze: molecular simulations unraveling the mysteries of Titan’s surface materials and liquids
No full textTitan, Saturn's largest moon, is a world of extraordinary astrobiological potential, characterized by its dense atmosphere, stable surface liquids, and seasonal weather cycles. With surface conditions that are cryogenic yet chemically rich, Titan hosts a complex interplay of organic molecules that could parallel early Earth's prebiotic chemistry. This unique environment, coupled with evidence of dynamic geological processes such as cryovolcanism and fluvial erosion, positions Titan as one of the most intriguing targets for exploring the possibility of life beyond Earth. However, a microscopic understanding of Titan's surface materials is missing, which is crucial for interpreting the complex interactions and processes occurring on its surface. This thesis aims to establish a foundational understanding of Titan's surface materials, particularly its cryominerals and hydrocarbon liquids, through state-of-the-art molecular simulations. We begin by focusing on the acetylene:ammonia (1:1) co-crystal, a representative Titan cryomineral. Our simulations reveal that at 90~K, the co-crystal exists in a dynamically disordered plastic phase, with ammonia molecules exhibiting rapid orientational disorder. We further demonstrate that under Titan’s cryogenic conditions, nuclear quantum effects can significantly accelerate this disorder by weakening intermolecular interactions. The implications of this phase transition on the mechanical properties of the co-crystal are explored, showing that dynamic disorder leads to anisotropic softening of the material's elastic constants. Additionally, we observe similar dynamic disorder in the acetonitrile (1:2) co-crystal, indicating that such disordered cryominerals are likely common on Titan’s surface. Along with the Titan's solids, this thesis also introduces dielectric corrected (DC) models for simulating liquid hydrocarbons, specifically methane and ethane, which are the primary components of Titan's lakes. These models incorporate fixed dipole moments to accurately represent the dielectric properties of the liquids, enabling the study of solvation phenomena that are crucial for understanding non-aqueous chemistry on Titan. The DC models are shown to provide a realistic description of solvation free energies for polar and charged species, thereby offering new insights into potential chemical processes in Titan’s unique environment. Throughout the thesis, local neural network interatomic potentials are employed to study Titan materials. In later sections, we investigate the limitations of these local models, particularly the consequences of neglecting long-range interactions. We examine the role of long-range interactions in orientational relaxations, solvation dynamics, and dielectric responses of polar liquids. Our findings indicate that while local models perform exceptionally well in predicting the structure and dynamics of bulk liquids, explicit long-range electrostatic interactions become critical at interfaces. Finally, we discuss the broader implications and future extensions of our work for understanding the surface and subsurface processes on Titan, as well as the potential for prebiotic chemistry. The insights gained from this thesis lay the groundwork for future investigations into the chemical, mechanical and thermodynamic aspects of Titan's materials, offering valuable predictions for upcoming missions such as NASA's Dragonfly rotorcraft, set to explore Titan in the 2030s.Ph.D.Includes bibliographical reference
Design and applications of porphyrin based nanocages for modelling reactivity inside and outside of porous materials
No full textThe incorporation of molecular catalysts into nanoporous materials, such as metal- and covalent organic frameworks, is likely to alter the outer sphere surrounding the active sites. To model the reactivity that occurs within these unique microenvironments, nanocages offer precise mechanistic insight and possess the rigid structure and porosity found within extended frameworks. In this work, a cubic M8L6 nanocage was used to encapsulate a CoII tetra-pyridyl porphyrin in order to provide a protective environment for a light sensitive CoIII-CH3 bond. Then, the development of a new covalent nanocage which is capable of binding fullerenes is discussed, along with an investigation on the electronic properties of the host-guest complexes. In the last chapter, the covalent nanocages are metalated with first row transition metals and used as heterogeneous CO2 electroreduction catalysts. Upon incorporation of fullerenes into the metalated cages to block the interior sites, the performance was enhanced for Fe analogues and relatively unperturbed when using Co, suggesting the hydrophobic pore of this nanocage may not be suitable for carrying out electrochemical transformations in aqueous conditions.Ph.D.Includes bibliographical reference
Innovative deep learning models and applications for diverse statistical problems
No full textThis thesis explores advanced methodologies and applications of deep learning models in various problems in statistics, including functional data analysis, graph classification, and large-scale classification. Key contributions include the development of a novel approach of nonlinear functional regression that leverages dynamic attention mechanisms to effectively handle both single and multiple functional covariates observed on regular and sparse irregular time grids; the introduction of a novel graph neural network model based on connection-diffusion ordinary differential equations, effectively capturing intricate relationships in graph-structured data; and a new family of weighted loss function for large-scale classification tasks. These methodologies provide powerful tools for analyzing complex datasets, with substantial implications for fields such as healthcare, finance, and biochemistry. Furthermore, this thesis demonstrates the application of machine learning and deep learning models to solve complex problems in biochemistry, illustrating how these methods can be leveraged to analyze large datasets, model intricate biological systems, and predict molecular behaviors with high accuracy. The innovative techniques presented in this work push the boundaries of current research, opening new avenues for exploration and showcasing the potential of deep learning to address complex challenges, ultimately driving progress in both scientific and industrial fields.Ph.D.Includes bibliographical reference
Effects of supramolecular structure on the redox, host-guest, and catalytic reactivity of metalloporphyrin nanocages
No full textSupramolecular nanocages constructed with redox active metalloporphyrin units provide a highly tunable environment for the study structural and secondary sphere effects on metalloporphyrin reactivity. Independently, metalloporphyrin units are used for chemical transformations within homogeneous and heterogeneous systems, but the fine tuning of these systems provides inspiration towards new research areas. Porphyrin-based supramolecular cages provide the foundation for these tuning studies by exploring the effects of the supramolecular structure on metalloporphyrin reactivity and by implementing desirable functional groups via non-covalent host-guest binding. Herin, we present a study into the redox, host-guest, and catalytic properties of a porphyrin-based supramolecular nanocage, exploring the effects of supramolecular structure on the properties of the porphyrin units. Introduction of CoII ions into the porphyrin walls provided a redox active metal site for reactivity studies, showing the stable reduction and oxidation of these units. The nanocage structure affected the metalloporphyrins access to the oxidized state due to steric strain, requiring the use of a stronger oxidant not affected by steric limitations. This structural affect was further explored by the coordination of neutral and anionic guests to the metal sites in the porphyrin nanocage’s walls. Steric limitations prevented larger guests from interacting with the interior of the nanocage while enhancing the coordination of smaller species. The nanocage was also able to non-covalently coordinate anionic guests into two cationic pockets within the nanocage structure, leading to the incorporation of acidic guests with increased acidity due to host-guest encapsulation. Acidic protons encapsulated within the nanocage structure were stoichiometrically deprotonated using oxygen reduction reactions in order to explore the effects of increased acidity on the proton’s accessibility towards the reduction of oxygen gas.Ph.D.Includes bibliographical reference
Electronic transport and planar tunneling spectroscopy in twisted graphene structures
No full textGraphene, since its first discovery, has attracted countless attention from various fields due to its unique honeycomb structure and novel properties. Stacking two layers of graphene with a twist adds an additional degree of freedom to tune the electronic properties since it generates a new periodic pattern besides the atomic lattice. It was shown that the central moiré bands become extremely flat near the so-called “magic angle” ~1.1°. Strong correlation effects dominate the carrier behavior due to the suppression of kinetic energy and bandwidth, exhibiting unexpected phenomena like unconventional superconductivity, quantized anomalous Hall effect, Chern insulators, heavy fermion states, etc. In this dissertation, I will present studies of electronic properties in magic angle twisted bilayer graphene (TBG) using both electronic transport and planar tunneling spectroscopy. The transport results focus on the conducting behavior of charge carriers on the Fermi surface while the tunneling measurements study the evolution of flat bands from the perspective of energy, not limited to the Fermi energy.Magic angle TBG is an ideal platform for studying strong correlation systems owing to the suppressed kinetic energy and the narrow isolated central bands. TBG near magic angle tends to have much higher resistivity but weaker carrier density dependence at room temperature, which might be related to scattering from the moiré superlattice. From the magnetoresistance and Hall resistance at low temperatures and under perpendicular magnetic fields, we discovered quantized Hall conductivity following non-zero Chern numbers and phase transitions marked by the resetting of Hall carrier density. Our discoveries reveal the non-trivial topology of the degenerate quasiparticle flat bands and demonstrate possible symmetry-breaking ground states at each integer filling which can be facilitated by van Hove singularities.
With the optimized design of tunneling device structure, flat bands with 26 meV total bandwidth reveal a cascade of phase transitions. An additional gap opening right at filling of 2 below 4 K agrees with the symmetry-breaking insulating state observed in transport measurements. Prominent heavy to light fermion transitions (depicted by the inverse compressibility and effective mass extracted from the tunneling current) at each non-zero filling and solid-like local moment states (shown as plateaus in the high-resolution entropy data) support the coexistence of localized f electrons and mobile c electrons discussed in the recently introduced topological heavy fermion models and Kondo lattice models.
Unusual phenomena like carrier confinement and negative differential conductance were also observed in TBG tunneling devices. It is demonstrated that the band gap between flat bands and remote bands amplifies the confinement of carriers induced by potential variations. The negative differential conductance originates from the misalignment of moiré flat bands in different regions, in a similar way to flat Landau levels developed under magnetic fields.Ph.D.Includes bibliographical reference
Exploring how parent social media use and communication impact teen social media experiences, depressive symptoms, and suicidal ideation
No full textSocial media (SM) use is often described as an issue for adolescents, with less researchexamining how parents play a role in their teens’ SM use and experiences. Understanding parent
SM behaviors, such as using SM in front of their teens (phubbing) and communication with teens
around SM, may be integral to understanding teens’ SM experiences and subsequent impact on
mental health outcomes, including depression and suicidal ideation (SI). The current study
included 413 adolescents (14-17; Mage = 15.98; 56% White; 45% girls) who completed a selfreport
survey of positive and negative SM-related emotional experiences (ERSM), SM checking
frequency, depressive symptoms, and SI (past month, year, lifetime SI). Adolescents also
reported perceptions of parents’ SM use frequency and phubbing, communication with parents
about SM, and household SM rules. Path analyses were conducted to examine if perceived parent
SM phubbing and parent-teen SM communication were associated with teens’ ERSM, and
whether teens’ ERSM was associated with depressive symptoms and SI. Indirect path analyses
examined whether parental phubbing and communication impacted teen depression and SI
through both negative and positive ERSM. There were significant indirect effects from perceived
parental phubbing to depressive symptoms through negative ERSM (B=.21, p=.005), and from
perceived parent-child SM communication to depressive symptoms through positive ERSM (B=-
.09, p=.006). Parents should consider limiting their use of SM in front of and while talking to
their teens. They should also foster open, non-judgmental communication with their teens about
SM experiences, which may be protective against negative mental health outcomes.M.S.Includes bibliographical reference
Advancements in modeling crowd navigation from cognition to simulation to prediction
No full textThis dissertation advances autonomous systems and cognitive modeling by addressing key challenges in simulation and prediction technologies. We propose novel solutions across several areas: cognitive approaches, spatial cognition, crowd behavior prediction, and trajectory forecasting. Traditional autonomous navigation models often rely on omniscient assumptions, which, while efficient, fail to capture individualized behaviors. Our new framework incorporates cognitive architectures with memory systems, enabling agents to create and use cognitive maps, thus simulating more realistic and nuanced wayfinding behaviors. In spatial cognition, we develop a computational methodology to standardize the analysis of structure mapping tasks. This approach enhances the accuracy of assessing spatial relationships and allows for better comparisons between human and agent representations. For crowd behavior prediction, we introduce a deep learning framework that utilizes the novel representation CAGE and a modified SegNet architecture. This framework provides instant long-term predictions of crowd flow, effectively capturing dynamics over extended periods and adapting to new environments. Furthermore, we improve prediction accuracy and scalability with Fourier Isovists and the GeoInteractNet framework, which combines multi-scale attention networks with convolutional encoder-decoder networks. In the domain of human trajectory prediction, we present the A2X dataset and new evaluation metrics, including multiverse metrics, to address existing limitations and enhance model reliability. This work integrates cognitive modeling, crowd flow prediction, and trajectory forecasting, offering practical solutions and advancing the state-of-the-art in autonomous systems and spatial cognition. The subsequent chapters detail the methodologies, experiments, and results supporting these advancements.Ph.D.Includes bibliographical reference