119092 research outputs found
Sort by
Computational design of magnetic materials for energy applications
No full textAt present there is a growing demand for refrigeration and cooling applications, with an expected commensurate increase in energy consumption. In this sense, functional magnetic materials enable numerous technologically relevant applications, such as magnetic cooling, energy conversion and energy harvesting. By exploiting the magnetocaloric effect, magnetic refrigeration promises to be an efficient and environmentally friendly cooling technology, however, it requires materials with a slew of optimized magnetic and structural properties, such as a large adiabatic temperature change and good mechanical behaviour. On the other hand, magnetic materials can exhibit emergent transport properties, e.g., the anomalous Hall and Nernst effects, with the latter allowing for the realisation of transverse thermoelectric devices. Such transport effects can be attributed to the presence of topological band structure features that contribute to the Berry curvature, e.g Weyl points and nodal lines, with their distance to the Fermi energy being a key factor and subject to optimization.
High-throughput screenings based on density functional theory calculations have been used as a method to accelerate materials discovery, with the aim of screening for stable compounds with intriguing properties. However, the application of such studies to magnetocaloric materials is still lacking, with no overreaching design strategy to tackle first-order magnetic transitions. In this work, a high-throughput procedure was implemented and applied to two classes of intermetallic systems - MM'X alloys (M & M'=metals, X=main-group) and all-d-metal Heusler compounds, in order to screen for promising candidates for caloric applications. We identified trends in key features of these systems, such as stability and relative bonding strengths in MM'X alloys. While in the all-d-metal Heusler family, we discussed how d-d bonding affects the mechanical behaviour and structural preference in this material class. In order to tackle the critical aspect of the magnetocaloric behaviour, we implement a computational workflow that screens for the magnetic and structural states that enable a first-order transition and then describes the finite-temperature effects that drive it. Using the notion of the Curie temperature window within an ab initio framework, we inform the selection and optimization of MM'X and all-d-metal Heusler compounds with potential magneto-structural transitions. We further predict how such properties can be controlled through isostructural substitution, guided by the results of our high-throughput study. We further explored the anomalous Hall and Nernst effects in the all-d-metal Heusler systems that were labelled as stable, and discussed the dependence of such effects on electron filling.
The results presented here serve to enrich automated ab initio workflows, representing a systematic methodology for evaluating the thermodynamic properties of magnetic materials, while the generated databases of MM'X and all-d-metal Heusler compounds provide a possible starting point for future machine learning studies. Thus, this work opens up new possibilities for the discovery and design of functional magnetic materials
K‐Doping Suppresses Oxygen Redox in P2‐Na₀.₆₇Ni₀.₁₁Cu₀.₂₂Mn₀.₆₇O₂ Cathode Materials for Sodium‐Ion Batteries
No full textIn P2‐type layered oxide cathodes, Na site‐regulation strategies are proposed to modulate the Na⁺ distribution and structural stability. However, their impact on the oxygen redox reactions remains poorly understood. Herein, the incorporation of K⁺ in the Na layer of Na₀.₆₇Ni₀.₁₁Cu₀.₂₂Mn₀.₆₇O₂ is successfully applied. The effects of partial substitution of Na⁺ with K⁺ on electrochemical properties, structural stability, and oxygen redox reactions have been extensively studied. Improved Na⁺ diffusion kinetics of the cathode is observed from galvanostatic intermittent titration technique (GITT) and rate performance. The valence states and local structural environment of the transition metals (TMs) are elucidated via operando synchrotron X‐ray absorption spectroscopy (XAS). It is revealed that the TMO₂ slabs tend to be strengthened by K‐doping, which efficiently facilitates reversible local structural change. Operando X‐ray diffraction (XRD) further confirms more reversible phase changes during the charge/discharge for the cathode after K‐doping. Density functional theory (DFT) calculations suggest that oxygen redox reaction in Na₀.₆₂K₀.₀₃Ni₀.₁₁Cu₀.₂₂Mn₀.₆₇O₂ cathode has been remarkably suppressed as the nonbonding O 2p states shift down in the energy. This is further corroborated experimentally by resonant inelastic X‐ray scattering (RIXS) spectroscopy, ultimately proving the role of K⁺ incorporated in the Na layer
Determination of Chain Transfer Constants of Alcohols for High‐Pressure Polyethylene Polymerization
No full textIn low‐density polyethylene (LDPE) production, low‐molecular‐weight species are used as chain transfer agents (CTAs). They reduce the degree of polymerization, which has an impact on the polymer properties, such as processability. To specifically adjust the product properties, a detailed understanding of the transfer kinetics to CTAs is indispensable. Thus, the transfer activity of various primary, secondary, and tertiary alcohols as well as diols was investigated experimentally between 120 °C and 250 °C at 2000 bar. The results were evaluated using the Mayo and chain length distribution (CLD) method, and the latter was found to give more reliable results. It was observed that the hydroxy group systematically increases transfer activity compared to alkanes. Additionally, a linear increase of transfer activity with the number of hydrogen atoms is found
Finite sections of truncated Toeplitz operators
No full textWe describe the C*-algebra associated with the finite sections discretization of truncated Toeplitz operators on the model space K²ᵤ where u is an infinite Blaschke product. As consequences, we get a stability criterion for the finite sections discretization and results on spectral and pseudospectral approximation
Light storage by electromagnetically induced transparency for one second at the level of a single photon in Pr3+:Y2SiO5 prepared with multiple frequency ensembles
No full textEfficient and long-term storage of quantum information encoded in single photons is crucial for applications of optical quantum technologies, e.g. quantum repeaters in communication networks. Obviously, the maximal storage time is an important benchmark for such memories, as it defines the distances covered by the network or the applicability of quantum communication protocols therein. In this paper, we present the implementation of an optical memory driven by electromagnetically induced transparency, permitting the storage of weak coherent pulses containing on average a single photon with a signal-to-noise ratio of 1.3(3) for a long storage time of one second. To achieve this goal, we apply decoherence control by static magnetic fields and robust dynamical decoupling sequences to prolong the coherence time in a rare-earth ion doped crystal to 14 s. A novel optical preparation scheme serves to increase the optical depth of the medium, which enables a light storage efficiency of 12.7(5)% at the single photon level and one second storage time
A Data-Driven Framework for Car-in-the-Loop Test Bench Control and Optimization
No full textThe mega-trends in automotive industry, from advanced driver assistance systems to electrification and autonomous driving, are driving the further development of the "Road-to-Rig" concept. This results in the emergence of the multi-dynamic complete vehicle test benches, which simulate dynamic driving scenarios in different degrees of freedom simultaneously under laboratory conditions. The Car-in-the-Loop (CiL) concept, which
has its advantages in terms of compactness and modularity, belongs to this class of test benches.
For the purpose of proof-of-concept, a quarter-vehicle CiL prototype was built in the laboratory of the Institute for Mechatronic Systems at the Technical University of Darmstadt. Previous work on the CiL control demonstrates the potential of the model-based controllers for the control of the test bench longitudinal dynamics. However, the control strategy that is implemented results in undesired oscillations on the side shaft of the vehicle under test (VUT) during dynamic driving scenarios that capture the tire slip and load change phenomena. Moreover, the drive torque from the VUT, which is not necessarily available in practice, is assumed known.
To address these issues, a data-driven framework called AI4CiL (A Data-drIven framework for Car-in-the-Loop control and optimization) is proposed in this work. It combines the test bench control with machine learning techniques. In AI4CiL, different model-based controllers that have the potential to improve the reference tracking quality and disturbance rejection behaviour are integrated. A generalized performance evaluation function is suggested, serving as the objective function for the core algorithm. To be specific, in AI4CiL, a Gaussain Process provides a model that maps the optimization variables to the
performance evaluation function. Bayesian Optimization utilizes this model to acquire the next iteration location which is supposed to maximize the performance improvement.
In this work, a variety of AI4CiL applications are implemented for the CiL control and optimization. Among others, AI4CiL is used to learn performance-driven models, which are optimized directly under closed-loop conditions with the aim to improve the control performance. Furthermore, safeOpt-type algorithm is integrated into AI4CiL to
safely optimize the controllers directly on the CiL prototype. The optimized controllers are experimentally validated with dynamic driving scenarios on the CiL prototype. It is demonstrated that different controllers are best suitable for different scenarios. For example, the LQG controller shows the best disturbance rejection behaviour, without inducing oscillations on the side shaft of the VUT. Finally, all the controllers demonstrate good reproducibility and robustness by performing tests on the CiL prototype at different steering positions.
In summary, with the help of AI4CiL, the control performance of the CiL concept for dynamic driving scenarios is significantly improved. Further development of controllers or optimization algorithms can be easily integrated into the existing structure. In addition, due to its practicability and flexibility, the proposed framework AI4CiL can be transferred for the control and optimization of other multi-dynamic complete vehicle test benches
Simplified Object Detection for Manufacturing: Introducing a Low-Resolution Dataset
No full textMachine learning (ML), particularly within the domain of computer vision (CV), has established
solutions for automated quality classification using visual data in manufacturing processes.
Object detection as a CV method for quality classification provides a distinct advantage
in enabling the assessment of items within the manufacturing environment, regardless of
their location in images. However, substantial challenges remain regarding labeled data
availability in manufacturing contexts, training examples, data imbalance, and the complexity
of incorporating these methods into real-world applications. Furthermore, real-world datasets
often lack adherence to FAIR principles, which limits their accessibility and interoperability,
especially for small- and medium-sized enterprises (SMEs) working to integrate object
detection into their manufacturing processes. In this article, we present a low-resolution
640x640 dataset based on plastic bricks for object detection, featuring two quality labels to
identify minor surface defects as an example of quality classification. We analyze the dataset
using a YOLOv5 model on three different dataset sizes, while accounting for class imbalance,
to demonstrate the accuracy of an object detection model in a simple manufacturing use
case. The mean Average Precision [email protected] for correctly identifying instances in our
testing dataset ranges from 0.668 to 0.774, depending on dataset size and class imbalance.
While our focus is on demonstrating object detection with low-resolution images and limited
data availability, the generated data and trained model also adhere to FAIR principles.
Therefore, these resources are made available with proper metadata to support their reuse
and further investigation into object detection tasks for similar quality classification use cases
in manufacturing
Coordination of Al(C₆F₅)₃ vs. B(C₆F₅)₃ on group 6 end-on dinitrogen complexes: chemical and structural divergences
No full textThe coordination of the Lewis superacid tris(pentafluorophenyl)alane (AlCF) to phosphine-supported, group 6 bis(dinitrogen) complexes [ML₂(N₂)₂] is explored, with M = Cr, Mo or W and L = dppe (1,2-bis(diphenylphosphino)ethane), depe (1,2-bis(diethylphosphino)ethane), dmpe (1,2-bis(dimethylphosphino)ethane) or 2 × PMe₂Ph. Akin to tris(pentafluorophenyl)borane (BCF), AlCF can form 1 : 1 adducts by coordination to one distal nitrogen of general formula trans-[ML₂(N₂){(μ-η¹:η¹-N₂)Al(C₆F₅)₃}]. The boron and aluminium adducts are structurally similar, showing a comparable level of N₂ push–pull activation. A notable exception is a bent (BCF adducts) vs. linear (AlCF adducts) M–N–N–LA motif (LA = Lewis acid), explained computationally as the result of steric repulsion. A striking difference arose when the formation of two-fold adducts was conducted. While in the case of BCF the 2 : 1 Lewis pairs could be observed in equilibrium with the 1 : 1 adduct and free borane but resisted isolation, AlCF forms robust 2 : 1 adducts trans[ML₂(N₂){(μ-η¹:η¹-N₂)Al(C₆F₅)₃}₂] that isomerise into a more stable cis configuration. These compounds could be isolated and structurally characterized, and represent the first examples of trinuclear heterometallic complexes formed by Lewis acid–base interaction exhibiting p and d elements. Calculations also demonstrate that from the bare complex to the two-fold aluminium adduct, substantial decrease of the HOMO–LUMO gap is observed, and, unlike the trans adducts (1 : 1 and 1 : 2) for which the HOMO was computed to be a pure d orbital, the one of the cis-trinuclear compounds mixes a d orbital with a π* one of each N₂ ligands. This may translate into a more favourable electrophilic attack on the N₂ ligands instead of the metal centre, while a stabilized N₂-centered LUMO should ease electron transfer, suggesting Lewis acids could be co-activators for electro-catalysed N₂ reduction. Experimental UV-vis spectra for the tungsten family of compounds were compared with TD-DFT calculations (CAM-B3LYP/def2-TZVP), allowing to assign the low extinction bands found in the visible spectrum to unusual low-lying MLCT involving N₂-centered orbitals. As significant red-shifts are observed upon LA coordination, this could have important implications for the development of visible light-driven nitrogen fixation
The State of Robo-Advisory Design: A Systematic Consolidation of Design Requirements and Recommendations
No full textAlthough robo-advisors offer potential benefits for enhancing investment decisions, financial decision-makers remain reluctant to utilize advice from robo-advisors, a form of artificial intelligence designed to convey newfound investment insights in a particularly intuitive way. To increase robo-advice utilization, numerous scholars have investigated various facets and design features aimed at making robo-advisors more appealing to investors. However, these proposed designs often focus only on specific aspects and are spread across various technological and domain contexts. Despite some overlapping ideas, there is little consistency between them. This has led to incoherent notions of what constitutes effective financial robo-advisor design among scholars and practitioners. To address this gap, we conduct a systematic literature review to synthesize existing knowledge. Based on 14 years of research, our study identifies six categories of design requirements and eight categories of design recommendations for robo-advisory. Our structured juxtaposition and synthesis of prevalent robo-advisor requirements and recommendations facilitate holistic research and practical implementations for viable financial robo-advisory systems
Konzeptionierung und Anwendung einer Methodik zum automatisierten Ableiten eines statischen Umgebungsmodells aus hochgenauen Karten für automatisiertes Fahren
No full textIm Rahmen des Projekts AUTOtech.agil haben sich 21 Universitäten und Unternehmen mit dem Ziel zusammengeschlossen, das Mobilitätssystem der Zukunft in Form einer offenen Architektur zu entwerfen. Ein Bestandteil des Projekts, der am Fachgebiet Fahrzeugtechnik der Technischen Universität Darmstadt erarbeitet wird, ist die Entwicklung einer Niedriggeschwindigkeitsfunktion (NGF). Diese soll in der Lage sein, verschiedene Anwendungsfälle im Niedriggeschwindigkeitsbereich mit einer entkoppelten Struktur von geringer Komplexität zu realisieren.
Neben einer eigenen Sensorik ist auch eine globale Trajektorienplanung ein Bestandteil der NGF. Diese berechnet übergeordnete Trajektorien in unstrukturierten Umgebungen wie beispielsweise Parkplätzen oder verkehrsberuhigten Bereichen. Um die unstrukturierten Umgebungen in einer Form abzubilden, die für die Trajektorienplanung geeignet ist, wird im Rahmen dieser Arbeit eine automatisierte Methode entwickelt und implementiert, die Informationen einer hochgenauen Karte verwendet und aus ihnen ein statisches Umgebungsmodell generiert.
Basierend auf den Anforderungen eines graphbasierten Trajektorienplaners und denen einer unstrukturierten Umgebung wird eine semantische Rasterkarte als Darstellungsform ausgewählt. Der gesamte Raum der Umgebung wird dabei abgebildet und kategorisiert, um auch Ausnahmesituation wie beispielsweise Parkvorgänge zu realisieren. Als hochgenaues Kartenformat zur Implementierung des Konzepts, wird das in der Forschung weit verbreitete Format Lanelet2 gewählt. Lanelet2 bietet neben flexiblen Darstellungsmöglichkeiten für unstrukturierte Umgebungen auch einen Quellcode, der frei verfügbar ist und somit die Umsetzung eigener Projekte erleichtert.
Implementiert wird die Methode zur automatisierten Ableitung eines statischen Umgebungsmodells Mithilfe der Programmiersprache Python. Zur Interpretation der Informationen der hochgenauen Karte werden bekannte Elemente anhand einer Zuordnungsliste kategorisiert. Unbekannte Elemente werden erfasst und anhand einer manuellen Abfrage ebenfalls den Kategorien zugeordnet. Abschließend werden sämtliche Elemente auf ein zweidimensionales Raster übertragen, das in Form einer Pixelmatrix ausgegeben wird.
Getestet wird die implementierte Methode zunächst mit selbst erstellten hochgenauen Karten, die sich an den Anwendungsfällen Automated Valet Parking und Verkehrsberuhigter Bereich orientieren. Im Anschluss erfolgen Tests mit fremd erstellten Karten sowie mit einem exemplarischen graphbasierten Trajektorienplaner. Dabei wird gezeigt, dass die Methode in der Lage ist, aus beliebigen hochgenauen Lanelet2-Karten ein Umgebungsmodell zu generieren, das für die globale Trajektorienplanung der NGF geeignet ist. Eine wesentliche Erkenntnis der Untersuchungen ist, dass die Qualität der hochgenauen Karten maßgeblichen Einfluss auf die Qualität des abgeleiteten Umgebungsmodells besitzt