DSpace@RPI (Rensselaer Polytechnic Institute)
Not a member yet
6809 research outputs found
Sort by
Helical vasculogenesis driven by cell chirality
No full textDecember 2023The morphogenesis of tubular structures is well-known for its crucial roles in development and disease, such as the formation of the cardiovascular system, which is among the first to exhibit left-right (LR) asymmetry. Nevertheless, the underlying mechanism governing this intricate phenomenon remains largely unexplored, leaving a significant knowledge gap, particularly in the context of recreating it in the well-controlled engineering systems. Leveraging advanced microfluidics and image processing, we present compelling evidence of the spontaneous emergence of helical endothelial tubes exhibiting a robust right-handedness governed by the inherent cell chirality, independent of vascular flow profile and substrate stiffness. To strengthen our findings, we also identify a consistent bias towards the same chirality in mouse retinal and aortic vascular tissues. Intriguingly, manipulating the chirality of endothelial cells through the administration of small-molecule drugs produces a dose-dependent reversal of the handedness in the engineered vessels, accompanied by non-monotonic changes in vascular permeability. Moreover, by probing into the biomechanical factors associated with this phenomenon, we demonstrate that the substrate surface curvatures and tissue fluidity play important roles in its regulation. In summary, our study unravels a novel mechanism underlying vascular chiral morphogenesis, shedding light on the broader implications and distinctive perspectives of tubulogenesis within biological systems.Ph
ChatBS: An Exploratory Sandbox for Bridging Large Language Models with the Open Web
No full textThe recent widespread public availability of generative large language models (LLMs) has drawn much attention from the academic community to run experiments in order to learn more about their strengths and drawbacks. From prompt engineering and fine-tuning to fact-checking and task-solving, researchers have pursued several approaches to try to take advantage of these tools. As some of the most powerful LLMs are ``closed'' and only accessible through web APIs with prior authorization, combining LLMs with the open web is still a challenge. In this evolving landscape, tools that can facilitate the exploration of the capabilities and limitations of LLMs are desirable, especially when connecting with traditional web features such as search and structured data. This article presents ChatBS, a web-based exploratory sandbox for LLMs, working as a front-end for prompting LLMs with user inputs. It provides features such as entity resolution from open knowledge graphs, web search using LLM outputs, as well as popular prompting techniques (e.g. multiple submissions, ``step-by-step''). ChatBS has been extensively used in Rensselaer Polytechnic Institute's Data INCITE courses and research, serving as key tool for utilizing LLMs outputs at scale in these contexts
Data-driven process monitoring and control for smart manufacturing
No full textDecember 2023Smart manufacturing (SM) is a new paradigm that uses information from equipment and sensors to make decisions that improve productivity, flexibility, and cost of running manufacturing processes. Attaining SM requires advanced monitoring and control systems to analyze real-time data and make decisions that drive the manufacturing process's overall goal. This dissertation focused on developing advanced process monitoring and control algorithms for SM applications. Data-driven process monitoring methods include Statistical and Artificial Intelligence (AI) approaches. Statistical methods often assume a linear behavior, which is not valid in many manufacturing processes. AI-based methods have been proposed to address the limitations of statistical methods. Another challenge is the availability of actual manufacturing data. Most of the proposed methods in the literature use the simulated Tennessee Eastman Process (TEP). We developed a novel AI-based method called Probabilistic bidirectional recurrent network (PBRN) for process monitoring using real plant data. Our model uses recurrent neural networks to learn long-term dependencies in the process data. We demonstrate the superiority of our model by comparing its performance with state-of-the-art statistical and AI-based methods for two case studies: the simulated TEP process and process data from an actual industrial air separation unit (ASU). We also reduce the computational requirements of the PBRN by introducing a shared parameter network (SPN), which uses a shared parameter space to learn relevant features for process monitoring. This architecture achieved a 75 percent reduction in model size and a 48 percent reduction in training time while maintaining a similar performance with the PBRN. Numerous studies have affirmed the competence of AI-based techniques in managing process monitoring and detecting faults. However, there is a noticeable divide between the promising results achieved in experimental setups and the real-world implementation of these AI strategies. Research indicates that only 13 percent of data science initiatives make it past the experimental phase to actual deployment in academic and industrial settings. Addressing this shortfall, we employ a shell and tube heat exchanger setup in the laboratory to demonstrate the deployment of AI models in real-time via a cloud-based manufacturing platform. This work is a hands-on approach to how AI models can move from conceptual studies to practical applications, driving toward an actual realization of Smart Manufacturing. The successful use of deep Reinforcement Learning (RL) in controlling hard-to-control dynamic systems such as the Cart-Pole, inverted pendulum, and Robotic arms has provided an opportunity to improve current process control techniques in the smart industry. RL algorithms can learn the optimal policies for controlling a system through repeated interactions. Model-free Rl methods require many repeated interactions, limiting their usage in critical manufacturing processes. Model-based RL methods have shown promising potential for reducing the required number of interactions to learn an optimal policy. We developed model-free and model-based RL algorithms for feedback control of the three-tank, quadruple-tank, and Van de Vusse systems. Our results show that model-based RL methods can track the setpoints of the dynamic systems studied.Ph
Low-power time-to-digital converters for high-precision measurement
No full textDecember 2023School of EngineeringTime-to-digital converters (TDCs), which converts time delays into digital signals, have garnered significant interest for their diverse applications in fields such as high-energy nuclear physics, time-of-flight (ToF) sensors, time-domain analog-to-digital converters (TD-ADCs), and all-digital phase-locked loops (ADPLLs). The performance of a TDC is primarily measured by its resolution, range, sampling rate, and power consumption. To simultaneously achieve a fine resolution and a long range, or a high dynamic range, in an area and power efficient manner, hierarchical architectures with the potential to combine the advantages of several different approaches has been studied. This thesis investigated three different hierarchical designs implemented in technologies ranging from 350nm CMOS to 45nm SOI, each suitable for its specific applications. This work first presents a hierarchical ADC-assisted TDC with reconfigurable resolution. The reconfigurable resolution and range are achieved by adjusting reference currents in the time-to-voltage converter (TVC) and the reference voltages in the ADC. The proposed resolution-reconfigurable approach combined with a two-step hierarchical architecture can be employed in a wide range of applications with different spatial range and resolution requirements. Fabricated using a 350nm CMOS process with a core area of 0.15mm², prototype chips yielded a resolution of 39ps with a 100MHz reference clock or 78ps with a 50MHz reference clock. In both cases, the measurement rate is 384kS/s while consuming less than 6.7mW from a 3.3V supply. Secondly, a multi-channel 4-tier TDC design combining gated-ring oscillators (GRO) coarse measurement stage, time amplifier, and 2D vernier fine measurement stage designed an simulated in 90nm SOI SiPh process. A dual-counter correction scheme is proposed to address the parallel-output-misalignment (POM) error in multi-phase clock based TDCs. The finer two tiers employ time amplifiers and 2D vernier lines to measure the residual signal, achieving a sub-gate-delay resolution while keeping a high conversion rate. Post-layout extracted simulation on the proposed TDC design shows a 2ps LSB size, while consuming 5.11-mW per additional channel from a 1.2V supply when operating at the maximum sampling rate of 500MS/s. Compared to state-of-the-art TDC designs, the proposed architecture shows an improvement in quantization step, conversion time, and dynamic range. The idea of multiphase-clock-based multichannel coarse measurement is further explored in a DLL-based TDC implemented in 45nm SOI technology. Lastly, a hierarchical pipeline TDC in 45nm SOI technology, optimized for high-speed and high-precision applications, is introduced. A novel analytical model for the cross-coupled time amplifier (TA) in the pipeline TDC is formulated. Based on this model, a gain calibration scheme is proposed. To validate the time amplifier analysis, a hierarchical TDC with pipeline fine measurement is designed and fabricated in 45nm SOI technology. With look-up table correction, measurement results of the TDC demonstrate a resolution of 0.95ps, a range of 0.8ns, and a DNL/INL range of 2.14 LSB and 2.13 LSB, respectively. The device operates on 8.851mW from a 1.0V supply at a sampling rate of 120MHz, and it achieves a maximum sampling rate of at least 250MHz, highlighting its capability to simultaneously deliver high speed and fine resolution.Ph
Factors controlling the catalytic activities of layered two-dimensional semiconducting fe and mn minerals
No full textDecember 2023Oxides and oxyhydroxides of iron and manganese, two of the most abundant elements on Earth, are ubiquitous in many geological settings such as oceans, lakes, soils, and sediments, where they perform many important biogeochemical functions. Among these various oxides, naturally formed two-dimensional layered minerals, such as birnessite (MnO2) and green rust (Fe(II)Fe(III)OOH), are powerful environmental catalysts for many species such as water, metal cations and organic pollutants, and play an important role in many biogeochemical cycles. Although the reactivity of these minerals has been well studied from the perspectives of crystal structure and composition, the electronic properties of these semiconducting minerals and their role in affecting the catalytic reactivity is not been well explored, which is the overarching goal of this dissertation. Studies were carried out with a wide variety of natural and synthetic birnessite samples. In the first study, the factors responsible for its high oxidation activity are evaluated across a series of birnessites with H, Li, Na, K, Cs, Ca, and Mg interlayer cations as well as other MnOx polymorphs. The results show that the oxidation activity of the cation-exchanged birnessite decreases in the order Mg > Cs = K > Na > Ca > Li > H. Their high reactivity is due to their unusually high electron affinity (∼5.9 eV), which is the highest among all known functional oxides and sulfide in aqueous solution. Both the band gap (Cs > Mg > K > Na > Ca > Li > H) and the electron affinity (Na > Cs > K > Mg > Ca > Li > H) are strongly affected by the nature of the interlayer cations and their coordinating water molecules. Analysis shows that the band gap increases with increase in the interlayer spacing, while the electron affinity increases with the relative Mn(III)/Mn(II) concentrations within birnessite. The high electron affinity values place the band edges of the birnessite-type structure well below the redox potential of most thermodynamically stable cations and water, thereby enabling spontaneous oxidation. Other tunnel-like and spinel MnOx polymorphs undergo slow phase transitions driven by Mn(II) lattice dissolution to form layered birnessite structures. The results reveal another unusual feature of manganese compounds specifically selected by nature for important biogeochemical functions. Birnessite, the closest natural analogue of Mn4CaO5 in the photosystem-II cluster, is also an important model compound for the development of biomimetic electrocatalysts for water oxidation reactions. Our second work reports the mechanism of formation of key Mn(III) intermediates achieved by studying the effect of several electrolyte anions and cations on the catalytic efficiency of birnessite. In situ Spectro-electrochemical measurements indicate that the activity is controlled by a dynamic solution-oxidation process in which Mn(III) is formed by oxidation of unstable uncomplexed Mn(II) in a manner similar to that of reversible shuttle between birnessite lattice and electrolyte: Photoactivation in Photosystem II. The role of electrolyte cations with different ionic radii and hydration strengths is to control the interlayer spacing, while electrolyte anions control the degree of deprotonation of complexed Mn(II) in the lattice. Both in turn control the shuttling efficiency of uncomplexed Mn(II) and its subsequent electrooxidation to Mn(III).
The result of our third study shows that the high oxidation activity of the various birnessites is linked to their high electron affinity, which in turn is modulated by various aqueous anions and cations through their dynamic influence on the dissolution-controlled structure Mn(III)/Mn(II). Inorganic cations with different ionic radii and hydration strengths control the interlayer spacing of the base layer, while inorganic and organic aqueous anions control the degree of Mn(II) and Mn(III) complexation. Their combined effect is thought to control the efficiency of manganese dissolution from the crystal, thereby modulating the activity and band structure of birnessite.
Finally, we evaluated the nature of ROS generated by photocatalytic excitation of various Fe and Mn oxides/hydroxides compared to photoactive anatase (TiO2) and clay mineral. Results show that Fe and Mn oxides have contrasting properties. Fe minerals, especially Fe(II)Fe(III) oxyhydroxide (green rust), have the highest activity for H2O2 generation but only a low to negligible activity for OH*, while Mn minerals (triclinic birnessite) have the highest activity of OH* and a low to moderate activity for H2O2. The highest concentrations of ROS seen with these minerals are nearly 5 to 7 times higher than that seen with TiO2, which is due to their unique electronic band alignments of birnessite and green rust minerals with respect to other common semiconductors.Ph
Unmasking the associations of lifestyle behaviors with type 2 diabetes, alzheimer’s disease and bone quality
No full textDecember 2023Bone is a dynamic tissue that, dependent on the state of its hierarchical levels, acts to provide support, and functions as a metabolic organ. The metabolism, structure, and function of bone can be affected by systemic diseases, such as Type 2 Diabetes (T2D) and Alzheimer’s Disease (AD). There are many contributing factors to T2D and AD occurrence, with research suggesting that lifestyle, environmental, and genetic factors all play an important role. Clinical studies have shown that lifestyle habits such as high-fat diets (HFDs) and circadian rhythm disruption (CRD) contribute to increase chronic inflammation and to the development and progression of T2D and AD. While these ubiquitous environmental factors can independently impact bone development, their interactions can have complex effects on bone quality and skeletal fragility. Indeed, skeletal fragility is a severe comorbidity of both T2D and AD. Diabetic patients, despite observed normal to elevated bone mineral density (BMD), have an increased risk for fragility fractures. In contrast, AD patients exhibit an increased risk for fracture with reduced BMD. Clinical studies link AD and osteoporosis, yet there is a lack of understanding on the effects of AD on bone beyond bone density. This study posits that, despite the contrary effects in bone quantity, both T2D and AD contribute to modifications of the primary constituents of bone (e.g., hydroxyapatite mineral, type I collagen, and non-collagenous proteins) and can decrease bone quality and increase bone fragility. For example, the increase levels of oxidative stress and inflammation, resulting from hyperglycemia or the neurodegeneration cascade, can promulgate the accumulation of advanced glycation end-products (AGEs). AGEs can stiffen the organic matrix, disrupt collagen organization, and alter fibrillar sliding leading to reduced energy dissipation in bone.
To address the above questions, this study utilizes a variety of preclinical animal (mouse) models to establish the effects of T2D and AD on bone matrix and skeletal fragility. The findings of this study highlight the two major pathways to skeletal fragility in AD through alteration of bone quality: accumulation of AGEs in the organic matrix; and loss of mineralization and crystallinity, and decreased crystal size. Furthermore, because HFD and CRD are often implicated in T2D and AD, an additional mouse model is employed here to determine the effects of the combination of CRD and a HFD on bone quality, independently of T2D and AD. Here, this study demonstrated that CRD together with a HFD during development leads to hyperglycemia and subsequent poor bone quality. A new strategy to manage AGEs from bone as a method for mitigating its effects on bone quality is also presented. This study demonstrated a partial rescue for HFD-induced skeletal fragility in a T2D mouse model using a combination of in vivo and in vitro approaches. It establishes the causality between accumulation of AGEs in bone fragility and provides a plausible avenue for clinical translation. Finally, a translational aspect of these preclinical findings is investigated using participant data from a nationally available clinical database to further explore the multivariable connection between sleep behaviors and incidence of diabetes, bone fracture risk, and cognitive dysfunction. This study examined the impact of sleep duration and sleep quality and an individual’s FRAX scores using the NHANES 2013-2014 dataset, wherein there was a trend of higher FRAX scores of the hip and of other sites for major osteoporotic fracturs amongst individual who reported sleep disturbances in an unadjusted model but also in a minimally and a fully adjusted model for all potential covariates.Ph
Incorporating context into knowledge graph completion methods
No full textMay 2024School of ScienceWell structured and semantically rich resources, such as knowledge graphs (KGs) and ontologies, can support tasks like knowledge-driven predictions or recommendations. KGs in particular have gained attention for such purposes in recent years due to their increasingly large scale and availability across domains. Given the directed graph structure of KGs, expressed as (subject, predicate, object) triples connecting the subject and object entities via the predicate relation, Knowledge Graph Completion (KGC) methods can be employed to identify missing entities or links in the KG in order to solve high-level tasks such as in recommendation or prediction systems. Such methods are, in a sense, using the KG to produce “new” knowledge rather than simply querying the KG for existing facts. In understanding the knowledge captured in a KG, as well as producing meaningful new knowledge for tasks predictions and recommendations, it is crucial to understand the context in which our information exists. However, while various context-aware applications have been developed over the years, the concept of context tends to be poorly defined and represented. Context might encapsulate information that characterizes an entity, or it might more generally describe the situation in which an activity is taking place. Context could also refer to background knowledge which somehow influences the outcome of a task. The delineation between what is or is not context is often left unclear due to the lack of any one-size-fits-all definition -- indeed, we often fall into a circular definition where what is context depends on the context. Towards addressing the problem of context and its use in knowledge-driven Artificial Intelligence (AI) applications, in this thesis we explore the development of KGC methods for producing new knowledge with an eye towards context. Our main technical contribution are three novel KGC methods which we demonstrate and evaluate in three distinct problem areas -- personalized food recommendation, event forecasting, and tabular data management. These three problem areas present us with a variety of different challenges and considerations that must be made to successfully apply KGC to the task, limiting the applicability of existing methods. For our first contribution, in the domain of food recommendation, we introduce novel methods to identify viable ingredient substitutions through generating and embedding flow graph representations of recipe instructions. More generally, these methods can be applied to the task of modifying entities that are involved in procedural instructions, allowing us to produce “new” knowledge by modifying existing knowledge. In our second contribution, tackling the domain of event forecasting, we present a novel model to predict properties of yet-unseen events based on performing 2-hop link prediction in a causal event KG. Our approach provides inherent explainability and requires no training, and can be applied more generally to predict properties about unseen entities in KGs -- this task of inductive link prediction enables us to produce “new” knowledge by making predictions about the properties of new entities. Our third contribution, for the domain of tabular data management, is the development of a method to predict table joinability using a relation prediction model over a KG of table metadata. We present a novel pipeline to process textual metadata into embeddings which are further processed by a graph neural network model, enabling joinability prediction for any new table and its metadata. This task produces “new” knowledge by predicting the likelihood of a new property -- specifically, joinability between tables -- existing between two target entities. Across these three contributions, we discuss a common theme of utilizing context surrounding entities and subgraphs in KGs. Further, based on our learnings from each of the technical contributions, we put forth a new ontology for context in KGC methods which can be used to explicitly model what is context in such methods. Our ontology enables users to represent and relate key concepts which are relevant to a context-aware KGC method, and more generally lays a foundation for defining context in a context-dependent manner. We demonstrate how this ontology can be applied using competency question and our three technical contributions as examples, and examine how our model can be applied more generally to common classes of KGC methods. We conclude by discussing the broader impact and limitations of our research, as well as future directions for the topic of effectively utilizing context in knowledge-driven AI.Ph
Living cyanobacteria architecture: human, nature, and machine in an occupied envelope
No full textAugust 2024School of ArchitectureThe scarcity of resources and catastrophic effects of climate change call for ecologically conscious built environments. Human activities, especially in the building sector, have contributed to global greenhouse gas emissions, waste generation, and depletion of renewable materials. Post-Anthropocene architecture necessitates an alternative environmental system approach, which involves incorporating nature in design thinking and processes to mitigate the crisis. In today's techno-bio-mediated societies, humans, nature, and machines interact with and construct each other. This research design aims to investigate how a living building system, involving interactions between humans, nature, and machines, can address environmental challenges in the building sector, such as CO2 emissions, waste treatment, and the use of renewable materials. The study is material-specific and used Cyanobacteria, a living and renewable matter, as an integral component of the living building system. The research progressed at three scales. In scale A, the living system was introduced in a lab-scale photobioreactor in an Arduino-based growth chamber. In scale B, the results and calculations from previous scale were extended to a specific location to study real site-specific environmental conditions on a larger surface scale. The system was considered in the context of a case study for a high-rise hotel in New York City. In the last scale, broader social and ecological relationships were considered. This involved the study of material and surface in an extended scope of marginal spheres and occupied envelopes. Results of the first scale indicate that with a photoperiod of L: D = 12 h: 12 h, red color LED with approximately 275.7 µmol/m2/s, effective mixing and aeration, we could monitor the temperature, pH and TDS real-time through the written Grasshopper script, and maintain them at approximately 30.58°C and 10.97 respectively. TDS values indicated the absence of harmful contamination in the culture. In the next step, the photobioreactors were parametrically designed and optimized for the specific site climate and solar data. In the last scale, in an occupied envelope, marginal spheres with wind cavities and thermal cyanobacteria windows were introduced. The power and heat generation and CO2 elimination of the living system were calculated. In a 40-story hotel with a 10,200 ft2 surface area, we have 972 active photobioreactors. The total 12 small roof-top wind turbines generate 83.62 MWh of power. It is estimated that the living system generates 11,9207.45 KWh of biomass heat annually and is estimated to eliminate 47.04 tons of CO2 annually. In conclusion, integrating cyanobacteria as a living material into a building system is a complex and multiscalar analysis. The Arduino-based growth chamber is effective in optimizing, monitoring, and simplifying the growth and cultivation process. In the next scale, the parametric system can effectively optimize the configuration of the photobioreactors in the south and east angles of the hotel building. In the last scale, the living system potential in power and heat generation and CO2 elimination were evaluated. Heat generation through biomass and power generation through the small wind turbines help to lower the building HVAC load. The ultimate goal is to promote ecological consciousness by showcasing the entire living system and creating spatial variabilities through thermal regulation strategies. The system analysis at multiple scales eventuates the potential of the living cyanobacteria architecture to mitigate CO2 emissions and waste while generating renewable sources in building systems.M
Growth and properties of single transition metal nitride-carbide superlattices
No full textAugust 2024School of EngineeringTransition metal nitrides (TMNs) and carbides (TMCs) are desirable materials for applications such as hard coatings, diffusion barriers, and optical coatings due to their high intrinsic mechanical strength, hardness, and thermal and chemical stability. Superlattices, nanoscale laminate composites consisting of alternating layers of two materials, allow for mechanical property enhancements through local strain variations and shear modulus mismatch which result in interfacial dislocation pinning at specific bilayer periods. Nanoscale engineering of TMNs and TMCs into superlattices increases the appeal within the application of hard, wear resistant coatings without significant alterations to existing fabrication techniques. I have studied three sets of single transition metal nitride/carbide superlattices, including TiN/TiC, MoN/MoC, and VN/VC, in order to investigate their hardness and elastic modulus as a function of bilayer period and the effect of material selection on the mechanical properties of superlattices. Further investigations develop a fundamental understanding of the effects of lattice constant and shear modulus mismatch on the potential enhancements of superlattice mechanical properties beyond intrinsic material values. Additionally, I have studied three sets of TiN/TiC superlattice series grown with different crystalline quality and growth orientations to better understand the effect that these properties have on superlattice hardening.Rocksalt structured TiN(001)/TiC0.5(001) superlattices with bilayer periods from 1.5 to 30 nm are grown on MgO(001) substrates via DC reactive magnetron sputtering in alternating Ar/N2 and Ar/CH4 gas mixtures at 1100 C. Microstructural characterization confirm epitaxial TiN(001) and TiC0.5x(001) demonstrating cube-on-cube epitaxy with the MgO(001) substrate, such that TiN/TiC[100] || MgO[100]. Measured XRD data shows an average superlattice relaxed lattice constant ao = 4.25 Å. The TiN(001)/TiC0.5(001) superlattices demonstrate superlattice enhancement at = 6 nm of H = 34 GPa, 1.4 the hardness of pure TiN and 1.1 that of pure TiC. Elastic modulus also has an enhancement at the same bilayer period with E = 750 GPa increasing 1.7 above that of pure TiN and TiC. The enhancements are attributed to shear modulus mismatch and lattice constant mismatch of 1.3 % forming coherency strains, both of which inhibit dislocation motion at superlattice interfaces and improve the mechanical properties. Additional rocksalt structured TiN/TiC superlattice series are grown on MgO(001) and Al2O3(0001) with increased carbon content. Microstructural characterization of TiN/TiC on MgO(001) confirms the presence of both (001) and (111) orientations, with a preferred (001) orientation. Hardness initially increases with bilayer period, reaching a maximum at = 7.5 nm of H = 17.9 GPa before decreasing with further increasing bilayer period. TiN/TiC on Al2O3(0001) demonstrates strong (111) growth and exhibits no increase in hardness above the intrinsic values of the separate materials. Overall, polycrystalline TiN/TiC superlattices have a reduced hardness and modulus relative to epitaxial samples, and the (001) orientations have increased mechanical properties compared to (111) orientations.
1.2 m thick molybdenum nitride/carbide superlattices with bilayer period = 1.5 to 30 nm are grown on MgO(001) substrates by DC reactive magnetron sputtering at 800 ℃. XRD confirms rock-salt structured MoN/MoC on MgO(001) with approximately random grain orientations. Mechanical property measurements yield a maximum of H = 12 GPa and E = 255 GPa at = 15 nm. The measured superlattice hardness is 471% and 50% larger than that of measured pure MoN and MoC, respectively. The superlattice hardening is the result of an 85 GPa shear modulus mismatch between MoN and MoC, as well as a 3.2 % lattice mismatch which generates stress fields and prevents dislocation motion.
Vanadium nitride/carbide superlattices 1.5 m thick are also grown on MgO(001) substrates with bilayer period = 1.9 to 30 nm at Ts = 1000 ℃. XRD measurements confirm that the VN/VC films exhibit an epitaxial rock-salt structured matrix, with cube-on-cube epitaxy with the MgO(001) substrate, such that VN/VC[100] || MgO[100]. Results of XRD on VN/VC superlattices show a measured relaxed lattice constant ao = 4.158 Å. Additionally, XRD,RSM, and SEM indicate that the samples are predominantly epitaxial with misoriented grains present. Reciprocal space mapping indicates increasing in-plane compressive strain as a function of increasing bilayer period. XPS and EDS measurements confirm stoichiometric VC but reduced nitrogen content. The VN/VC superlattice films indicate a decreased hardness with increasing bilayer period, from H = 16.4 to 11.1 GPa, and elastic modulus E = 260 GPa that is constant as a function of bilayer period. Although the material system does have a shear modulus mismatch of 67 GPA, the superlattice hardness values do not exceed those of pure VN or VC likely due to a low lattice mismatch (1.0 %) which suppresses the typical mechanical property enhancement exhibited by superlattices.Ph
An automated modeling, meshing, and adaptive framework for tokamak plasma simulations.
No full textMay 2024School of EngineeringThe accurate representation of a problem domain in terms of a geometric model and its effective discretization into a mesh plays an important role in achieving higher-quality simulation results and better computational efficiency. In this work, an automated modeling and meshing infrastructure with adaptive mesh control is developed to support large-scale fusion plasma simulations. The goal of the presented work is to develop a framework that accurately constructs the tokamak geometric models and discretize these model such that they meet the simulation needs of specific simulation codes. The modeling and meshing procedures for two fusion plasma codes, XGC and M3D-C1, are presented.An automated modeling and meshing framework, TOMMS, was developed to support the magnetic field-aligned and one-element deep meshing requirements of XGC. Recent developments to extend this infrastructure to better support the near flux field following meshing requirements of the XGC are presented. The first extension is a procedure to effectively represent a general set of O-point and X-points configurations. The second extension is a procedure to decompose the geometric model that includes the tokamak wall, selected flux curves and the separatrix curves in a manner such that an appropriate set of mesh generation procedures can be applied to each sub-region. The third extension is the application of appropriate mesh generation procedures in each sub-region to create the desired final mesh. Also, the procedures to provide the mesh information needed for the efficient execution of XGC computational operations are presented.
The M3D-C1 code requires unstructured high-order triangular elements on the 2D poloidal plane for 2D simulations and 3D wedge elements which are constructed from the extrusion of 2D elements in the toroidal direction for 3D simulations. As part of work to extend M3D-C1 to a wide range of tokamak geometries, a generalized modeling and meshing infrastructure was developed to support the modeling of tokamaks with any configuration of physical features. The goal of this development is to support construction of model geometries with an arbitrary number of model loops and better control on mesh property settings on individual model entities. Moreover, a set of procedures to support a-priori mesh
modifications to generate the initial M3D-C1 meshes with desired resolution are presented. In M3D-C1 simulations, the plasma equilibrium evolves over time. To effectively simulate those changes in plasma during a simulation, the mesh needs to be adapted dynamically. A mesh adaptation approach driven by an SPR error estimator is developed. 2.5D mesh
adaptation procedure to extend 2D error-based mesh adaptation approach to 3D is also presented.Ph