256869 research outputs found
Sort by
Frequency reproducibility of solid-state thorium-229 nuclear clocks
No full textSolid-state thorium-229 (²²⁹Th) nuclear clocks are set to provide new opportunities for precision metrology and fundamental physics. Taking advantage of inherent low sensitivity of a nuclear transition to its environment, orders of magnitude more emitters can be hosted in a solid-state crystal compared with current optical lattice atomic clocks10. Furthermore, solid-state systems needing only simple thermal control11 are key to the development of field-deployable compact clocks. Here we explore and characterize the frequency reproducibility of the ²²⁹Th:CaF₂ nuclear clock transition, a key performance metric for all clocks. We measure the transition linewidth and centre frequency as a function of the doping concentration, temperature and time. We report the concentration-dependent inhomogeneous linewidth of the nuclear transition, limited by the intrinsic host crystal12 properties. We determine an optimal working temperature for the ²²⁹Th:CaF₂ nuclear clock at 196(5) K, at which the first-order thermal sensitivity vanishes. This would enable in situ temperature co-sensing using different quadrupole-split lines, reducing the temperature-induced systematic shift below the 10⁻¹⁸ fractional frequency uncertainty level. At 195 K, the reproducibility of the nuclear transition frequency is 220 Hz (fractionally 1.1 × 10⁻¹³) for two differently doped ²²⁹Th:CaF₂ crystals over 7 months. These results form the foundation for understanding, controlling and harnessing the coherent nuclear excitation of ²²⁹Th in solid-state hosts and for their applications in constraining temporal variations of fundamental constants
A Structural Approach to Query Optimisation for Efficient Join and Aggregate Processing
Get PDFJoin processing remains a fundamental challenge in database systems, particularly when intermediate results grow large. This problem is frequently encountered in analytical queries, which aggregate data over many relations, but output only a comparably small final result. Although the vast majority of user-generated queries are acyclic, and the foundational result by Yannakakis provides a solution for avoiding unnecessary computation of intermediate results in theory, structure-guided query evaluation remains absent from standard database systems. In this thesis, we work towards closing this gap from theory to practice.To explore the benefits and limitations of structure-guided query processing, we introduce a query-rewriting-based implementation of Yannakakis' algorithm, on top of different DBMSs. Additionally, we identify the class of zero-materialisation answerable (0MA) queries, which can be answered efficiently without materialising join results. Experimental evaluation shows that the rewriting approach can achieve significant speedups on hard instances but fails to outperform in most situations. These results motivate us to frame the problem as an algorithm selection problem -- we therefore apply machine learning to solve this problem, setting up a new data set of queries in the process, and verify that we can reliably decide when to apply the rewriting.Following up on these encouraging results, we go beyond the quite restrictive 0MA class, introducing guarded and piecewise-guarded aggregate queries, allowing us to cover a wide range of aggregate queries while avoiding materialisation. We integrate these optimisations into the query optimiser of Spark SQL by implementing logical optimisations and a new physical operator AggJoin. Through experimental evaluation, we show that this implementation speeds up aggregate processing on many queries of several benchmarks while never degrading performance. Furthermore, to cover arbitrary unguarded queries, where materialisation cannot be fully avoided, we extend the AggJoin to the GroupAggJoin operator, substantially reducing materialisation on these queries as well. By establishing a new benchmark for unguarded queries, we show that this approach successfully handles moderately unguarded queries.Having established efficient techniques for the processing of large classes of acyclic aggregate queries, we focus on the particularly challenging class of cyclic queries. To apply the standard approach of using decompositions to make cyclic queries acyclic successfully, it is clear that we require secondary optimisation objectives in addition to minimising the width of decompositions. To address this challenge, we introduce soft hypertree decompositions and the associated measure of soft hypertree width (shw). By avoiding the special condition of hypertree decompositions, we gain algorithmic flexibility while retaining the tractability of computing width-k decompositions. By developing an end-to-end pipeline, we achieve a seamless integration of these optimisations into DBMSs. Experimental evaluation on cyclic queries confirms that we can achieve significant improvements in decomposition quality, which translates to performance gains in practice
Nonlocal spatial correlations in an interacting atomic gas
No full textWe report in situ measurements of the nonlocal two-body spatial correlation function ⁽²⁾() in an ultracold Bose gas with strong interactions using photoassociation of ultra-long-range Rydberg molecules (ULRRMs) as a spatially selective probe. This technique provides access to ⁽²⁾() on length scales from 60 to 180 nm, bridging the gap between short-range and long-range correlation measurements. The large -wave scattering length of ⁸⁶Sr (=811₀) enables exploration of the universal regime where ⁽²⁾() is governed solely by two-body interactions. Comparison with a weakly interacting isotope, 84 Sr, highlights interaction-induced suppression of pair correlations. The measured correlation functions agree quantitatively with ab initio calculations based on the two-body reduced density matrix, without adjustable parameters. Our results establish Rydberg molecular photoassociation as a powerful tool for probing nonlocal correlations in situ in strongly interacting quantum gases
Reaction kinetics of recycling binder as substitutes in cement
Get PDFZement ist ein zentraler Bestandteil von Beton und weltweit einer der meistverwendeten Baustoffe. Zement stellt jedoch eine bedeutende Quelle von \ch{CO2}-Emissionen dar. Gleichzeitig fallen im Bau- und Abbruchsektor große Mengen an Abfällen an, deren Deponierung zunehmend als nicht nachhaltig gilt. Vor diesem Hintergrund untersucht diese Arbeit die Eignung von drei industriellen Recyclingmaterialien (RM1, RM2 und RM3) als teilweisen Ersatz für Portlandzement (CEM) und vergleicht ihre Leistungsfähigkeit mit etablierten Zementersatzstoffen wie Metakaolin und Kalksteinmehl sowie mit einem inerten Ersatzstoff wie Quarzsand. Die Reaktionskinetik der Recyclingmaterialien wurde durch kalorimetrische und rheologische Untersuchungen analysiert, während ihr Einfluss auf die Festigkeitsentwicklung anhand von Druckfestigkeitsprüfungen an Mörtelproben bewertet wurde. Ergänzend wurden Ausbreitmaß, Erstarrungsbeginn, Trocknungsschwinden sowie die Effekte unterschiedlicher Lagerungsbedingungen untersucht.Die kalorimetrische Untersuchung der reinen Recyclingmaterialien zeigt deutliche Unterschiede in deren Reaktivität: RM1 und RM2 setzen nur geringe Wärmemengen frei und verhalten sich weitgehend inert, während RM3 eine frühe Reaktivität und eine deutlich höhere Wärmefreisetzung aufweist. Diese frühe Reaktivität von RM3 im Kontakt mit Wasser wird durch SAOS-Messungen bestätigt, bei denen der Speichermodul früh und steil ansteigt, was auf die schnelle Ausbildung eines flokkulierten Netzwerks hinweist, während RM1 und RM2 nur geringe Änderungen zeigen und überwiegend inert bleiben. In Zementmischungen bewirkt RM3 eine Beschleunigung der anfänglichen Hydratationsreaktionen, führt jedoch gleichzeitig zu einer Verzögerung der Beschleunigungsphase der Zementhydratation. Im Gegensatz dazu verlangsamen RM1 und RM2 die Beschleunigungsphase der Zementhydratation nur in geringem Maße und zeigen insgesamt weniger ausgeprägte Effekte. Der Effekt von RM1 und RM2 auf die Zementhydratation ist mit dem Einfluss von Kalksteinmehl und Metakaolin vergleichbar.Mechanisch hängt die Festigkeitsentwicklung der CEM:RM-Mischungen stark vom Substitutionsgrad und den Lagerungsbedingungen ab. Bei einem hohen Ersatzstoffanteil von 50:50 sind die Druckfestigkeiten deutlich geringer, sodass die Materialien für Anwendungen mit niedriger Traglast geeignet sind. Eine Reduktion des Recyclingmaterialanteils auf 30 \% führt zu einer deutlichen Festigkeitssteigerung, wobei RM3 unter optimierten Bedingungen hohe Druckfestigkeiten erreicht, die mit denen von Metakaolin vergleichbar sind.Insgesamt zeigt die Arbeit, dass der teilweise Ersatz von Zement durch Recyclingmaterialien nicht nur die \ch{CO2}-Emissionen verringert und Primärressourcen schont, sondern auch die Entwicklung nachhaltiger Betonanwendungen ermöglicht. Besonders RM3 weist nach geeigneter Nachbehandlung ein großes Potenzial für Anwendungen mit höheren mechanischen Anforderungen auf.Cement is essential to concrete and ranks among the most heavily used building materials globally; yet, it is a major source of \ch{CO2} emissions. At the same time, construction and demolition activities generate vast quantities of waste that are increasingly unsustainable to landfill. In response, this thesis investigates the suitability of three industrial recycled materials (RM1, RM2, and RM3) as partial replacements for Portland cement (CEM), comparing their performance with well-known supplementary cementitious materials such as metakaolin and limestone powder, as well as an inert substitute like quartz sand. The focus is on reaction kinetics, determined through calorimetric and rheological measurements, as well as on the influence of the recycled materials on strength development, evaluated via compressive strength tests on mortar samples. Additional properties, including spread flow, Vicat measurements, drying shrinkage, and the effects of different curing conditions, were also examined.The calorimetry of the pure recycled materials highlights differences in reactivity: RM1 and RM2 release relatively little heat and appear largely inert, whereas RM3 exhibits early reactivity and a significantly higher heat release. Rheological measurements confirm this behaviour, with RM3 showing a rapid increase in the storage modulus , indicating the fast formation of a flocculated network, while RM1 and RM2 remain largely inert. When incorporated into cement, RM3 accelerates early hydration reactions but delays the main phase of cement hydration, whereas RM1 and RM2 have only minor effects and cause a smaller delay of the main phase. Overall, the influence of RM1 and RM2 on cement hydration is comparable to that of limestone powder and metakaolin.Mechanically, the performance of CEM:RM mixtures strongly depends on the substitution level and curing conditions. At a 50:50 volume ratio, compressive strengths are substantially lower than those of pure cement, suggesting suitability primarily for low load-bearing applications. Reducing the substitution to 30\% markedly improves strength, particularly for RM3, which, under optimised curing conditions, achieves compressive strengths comparable to those of metakaolin and shows promising overall mechanical performance.Overall, partial replacement of cement with recycled materials can significantly reduce \ch{CO2} emissions, conserve primary raw materials, and provide a sustainable alternative for concrete production. In particular, RM3 demonstrates the greatest potential for applications requiring higher mechanical performance following appropriate curing, highlighting its suitability for more demanding construction applications
Accelerating First-Principles Molecular-Dynamics Thermal Conductivity Calculations for Complex Systems
No full textAtomistic simulations of heat transport in complex materials are costly and hard to converge. This has led to the development of several noise-reduction techniques applicable to equilibrium molecular-dynamics (MD) simulations. We analyze the performance of those strategies, taking InAs nanowires as our benchmark due to the diverse structures and complex phonon spectra of these quasi-1D systems. We demonstrate how, for low-thermal-conductivity systems, cepstral analysis can reduce computational demands while still delivering accurate results that do not require discarding arbitrary parts of the data set. However, issues with this approach are revealed when treating high-thermal-conductivity systems, where the thermal conductivity is significantly underestimated. We discuss alternative methods to be used in that situation, relying on uncertainty propagation from independent simulations. We show that the contributions of the covariance matrix have to be included for a quantitative assessment of the error. The combination of these strategies with machine-learning interatomic potentials (MLIPs) provides an accelerated, robust workflow applicable to a diverse set of systems, as our examples using a highly transferable MACE potential illustrate
MHIT36: Extension to wall-bounded turbulence and scalar transport equation
Get PDFWe present an extended version of MHIT36, a GPU-tailored solver for interface-resolved simulations of multiphase turbulence. The framework couples direct numerical simulation (DNS) of the Navier–Stokes equations, which describe the flow field, with a phase-field method to capture interfacial phenomena. In addition, the transport equation for a scalar can also be solved. The governing equations are discretized using a second-order finite difference scheme. The Navier–Stokes equations are time advanced with an explicit fractional-step method, and the resulting pressure Poisson equation is solved using a FFT-based method. The accurate conservative diffuse interface (ACDI) formulation is used to describe the transport of the phase-field variable. Simulations can be performed in two configurations: a triply-periodic cubic domain or a rectangular domain of arbitrary dimensions bounded by two walls. From a computational standpoint, MHIT36 employs a two-dimensional domain decomposition to distribute the workload across MPI tasks. The cuDecomp library is used to perform pencil transpositions and halo updates, while the cuFFT library and OpenACC directives are leveraged to offload the remaining computational kernels to the GPU. MHIT36 is developed using the managed memory feature and it provides a baseline code that is easy to further extend and modify. MHIT36 is released open source under the MIT license
Medical support platform for melanoma analysis and detection based on federated learning
No full textAdvances in computer science and medicine have led to the emergence of artificial intelligence as a key tool in the medical and scientific fields. Its application in the diagnosis and treatment of diseases, such as cancer, has proven to be fundamental in improving early detection and saving lives. This article presents a proposal based on Deep Learning to develop a model capable of detecting melanomas in the skin from clinical images. The aim is to provide doctors with a tool to support early identification of this type of cancer, considering additional factors such as sun exposure and the patient's skin tone. To optimize diagnostic accuracy and prevent data silos, a collaborative learning technique called Federated Learning is implemented. The FL framework employs a weighted averaging algorithm for model aggregation, allowing locally trained models to contribute to a continuously improving global model without sharing patient data. Experiments show that the proposed federated model achieved an accuracy of 89.1% and a ROC AUC of 0.9251, demonstrating performance comparable to centralized training while preserving privacy. In addition, a web application is presented to manage and process the information efficiently, making it easier for doctors to consult and analyze the results
The Future of Rural Ensembles - Strategies of Continued Building and Reuse in the Vienna Woods
No full textWohnungsnot und Leerstand sind zwei widersprüchliche Phänomene, die das Spannungsfeld der heutigen Stadtentwicklung sichtbar machen. Im urbanen Kontext wird der Umgang mit diesem Thema zunehmend diskutiert und kritisch hinterfragt. Der ländliche Raum bleibt in dieser Debatte meist unbelichtet und findet in der Gesellschaft deutlich weniger Beachtung, obwohl Leerstand in vielen Regionen das Ortsbild prägt. Strukturelle Veränderungen, wie Abwanderung und demografischer Wandel, führen zum Verfall ländlicher Hofensembles und werfen Zukunftsfragen über die neue Rolle des ländlichen Raumes auf. Ausgehend von der städtischen Leerstandsproblematik in Wien, richtet sich mein Blick auf die Marktgemeinde Mauerbach am westlichen Stadtrand, in der ein Ensemble als Sinnbild für ungenutzte Potenziale steht. Obwohl die Marktgemeinde nur unweit des urbanen Geschehens liegt, bildet dessen Ortsmitte ein historisch gewachsenes Gefüge, in dem zahlreiche Räume ungenutzt bleiben. In den meisten Fällen führt der Verfall ungenutzter, jedoch intakter Gebäude zum Abriss und in Folge dessen zur Neubebauung. Dies ist nicht nur in Zeiten der Klimakrise und Ressourcenknappheit stark zu hinterfragen. Um eine nachhaltige Baukultur zu schaffen, muss das Potenzial des Bestands erkannt, gewertschätzt und genutzt werden. Gebaute Strukturen sind Rohstofflager und integrale Bestandteile des kulturellen und räumlichen Gefüges, die wieder zum Leben erweckt werden wollen. Vor diesem Hintergrund untersucht die vorliegende Diplomarbeit neue Perspektiven im Umgang mit obsolet gewordenen Hofensembles. Durch selektiven Rückbau und Bauteilwiederverwendung eröffnen sich neue Wege, um an Bestandsstrukturen weiterzubauen und die vor Ort vorhandenen Ressourcen neu zu interpretieren. Ein Blick in die praktische Baukultur zeigt die Relevanz der Typologie von Bauteillagern, welche eine mögliche Nutzungstransformation darstellen. Alte Materialien werden gesammelt, aufbereitet und anschließend wieder in den Kreislauf des Bauens zurückgeführt. Es zeigt sich für mich ein verborgener, vielschichtiger Möglichkeitsraum, der Schnittmengen zwischen Arbeiten, Leben und Wohnen bietet. Mauerbach als Ort der Begegnung zwischen Natur und gewachsener Infrastruktur entfaltet einen Spielraum für die Gemeinde und Wien. Anhand von Fragmenten des Ensembles stellt der Entwurf mögliche Konzepte für eine für die Gemeinde attraktive Transformation dar. In meiner Diplomarbeit beschäftige ich mich mit Ansätzen, die aufzeigen, dass oftmals mit minimalem Aufwand deutliche Veränderungen erzielt werden können. Durch das Weiterbauen vorhandener Substanz durch Bauteilwiederverwendung wird das Narrativ unserer Baukultur nachhaltig bewahrt und fortgeschrieben.Housing shortages and vacancy are two opposing phenomena that illustrate the tensions associated with current urban development. In an urban context, the handling of this issue is increasingly being discussed and critically questioned. Rural areas are largely ignored in this debate and receive significantly less attention from society, even though vacancy shapes the towncape in many regions. Structural changes, such as migration and demographic change, are leading to the decline of rural farm ensembles and raise future questions about the new role of rural areas. Starting from the urban vacancy problem in Vienna, my attention is directed towards the market town of Mauerbach on the western outskirts of the city, where an ensemble stands as a symbol of untapped potential. Although the market town is located not far from the urban centre of activity, its town centre is a historically grown structure in which numerous spaces remain unused.In most cases, the decay of unused but intact buildings leads to demolition and consequently new construction. This is not only to be questioned in times of climate crisis and resource scarcity. To create a sustainable building culture, the potential of existing buildings must be recognised, valued, and utilised. Built structures are raw material warehouses and integral components of the cultural and spatial fabric that want to be brought back to life. Against this backdrop, this diploma thesis investigates new perspectives in dealing with obsolete courtyard ensembles. Selective deconstruction and component reuse open up new ways for expanding existing structures and reinterpreting locally available resources. A look at practical building culture shows the relevance of the typology of building component warehouses, which represent a possible transformation of use. Old materials are collected, processed, and then reintroduced to the building cycle. For me, this reveals a hidden, multifaceted world full of possibilities that offers intersections between work, life, and living. Mauerbach, a place of encounter between nature and established infrastructure, unfolds a scope for the community and Vienna. Based on fragments of the ensemble, the design presents possible concepts for a transformation that is attractive to the community. In my diploma thesis, I explore with approaches that demonstrate how significant changes can often be achieved with minimal effort.By continuing to build on existing substance by reusing building components, the narrative of our building culture is sustainably preserved and perpetuated