arXiv subjects

Daniel Wolf

Publications and source records attributed to Daniel Wolf.

At least 19 recordsLinked to original sources

arXiv2026

Generation of Bloch Points with Controlled Spin Texture Using Geometrical Boundary Conditions

Bloch points are three-dimensional topological singularities in magnetization that play a key role in topological transformations of spin textures, such as skyrmion creation or annihilation. While topology often enforces the existence of Bloch points in confined geometries like cylindrical nanowires, deterministic control over their position and magnetic configuration remains challenging. Here we demonstrate the generation of Bloch points with controlled spin texture by engineering geometrical boundary conditions in three-dimensional nanomagnets. By introducing a chirality interface between two three-dimensional double-helix nanowires of opposite handedness, forming a kinked, non-collinear structure, we impose competing topological constraints that uniquely define the magnetization configuration surrounding the Bloch point. A saturating magnetic field nucleates head-to-head or tail-to-tail domain configurations at the chirality interface, producing a Bloch-point domain wall with deterministic polarity, circulation and helicity. This geometrical approach enables full three-dimensional control of Bloch point domain walls allowing deterministic engineering of their spin texture and its selective coupling to current-induced Oersted fields.

cond-mat.mes-hall↗

arXiv2025

Curvilinear magnonic crystal based on 3D hierarchical nanotemplates

Curvilinear magnetic nanostructures enable control of magnetization dynamics through geometry-induced anisotropy and chiral interactions as well as magnetic field modulation. In this work, we report a curvilinear magnonic crystal based on large-area square arrays of truncated nanospikes fabricated by conformal coating of 3D hierarchical templates with permalloy thin films. Brillouin light scattering spectroscopy reveals anisotropic band structure with multiple dispersive and folded Bloch-type dispersive spin-wave modes as well as non-dispersive modes exhibiting direction-dependent frequency shifts and intensity asymmetries along lattice principal axes. Finite element micromagnetic simulations indicate that curvature-induced variations of the demagnetizing field govern the magnonic response, enabling the identification of modes propagating in nanochannels and other localized on nanospike apexes or along the ridges connecting adjacent nanospikes. The combination of geometric curvature and optical probing asymmetry produces directional dependence of magnonic bands, establishing 3D hierarchical templates as a versatile platform for curvature-engineered magnonics.

cond-mat.str-el↗

arXiv2025

Your other Left! Vision-Language Models Fail to Identify Relative Positions in Medical Images

Clinical decision-making relies heavily on understanding relative positions of anatomical structures and anomalies. Therefore, for Vision-Language Models (VLMs) to be applicable in clinical practice, the ability to accurately determine relative positions on medical images is a fundamental prerequisite. Despite its importance, this capability remains highly underexplored. To address this gap, we evaluate the ability of state-of-the-art VLMs, GPT-4o, Llama3.2, Pixtral, and JanusPro, and find that all models fail at this fundamental task. Inspired by successful approaches in computer vision, we investigate whether visual prompts, such as alphanumeric or colored markers placed on anatomical structures, can enhance performance. While these markers provide moderate improvements, results remain significantly lower on medical images compared to observations made on natural images. Our evaluations suggest that, in medical imaging, VLMs rely more on prior anatomical knowledge than on actual image content for answering relative position questions, often leading to incorrect conclusions. To facilitate further research in this area, we introduce the MIRP , Medical Imaging Relative Positioning, benchmark dataset, designed to systematically evaluate the capability to identify relative positions in medical images.

cs.CV↗

arXiv2025

Inducing ferromagnetism by structural engineering in a strongly spin-orbit coupled oxide

Magnetic materials with strong spin-orbit coupling (SOC) are essential for the advancement of spin-orbitronic devices, as they enable efficient spin-charge conversion, complex magnetic structures, spin-valley physics, topological phases and other exotic phenomena. 5d transition-metal oxides such as SrIrO3 feature large SOC, but usually show paramagnetic behavior due to broad bands and a low density of states at the Fermi level, accompanied by a relatively low Coulomb repulsion. Here, we unveil ferromagnetism in 5d SrIrO3 thin films grown on SrTiO3 (111). Through substrate-induced structural engineering, a zigzag stacking of three-unit-cell thick layers along the [111] direction is achieved, stabilizing a ferromagnetic state at the interfaces. Magnetotransport measurements reveal an anomalous Hall effect below ~30 K and hysteresis in the Hall conductivity below 7 K, indicating ferromagnetic ordering. X-ray magnetic circular dichroism further supports these results. Theoretical analysis suggests that the structural engineering of the IrO6 octahedral network enhances the density of states at the Fermi level and thus stabilizes Stoner ferromagnetism. This work highlights the potential of structurally engineered 5d oxides for spin-orbitronic devices, where efficient control of SOC-induced magnetic phases by electric currents can lead to lower energy consumption and improved performance in next-generation device technologies.

cond-mat.mtrl-sci↗

arXiv2024

Less is More: Selective Reduction of CT Data for Self-Supervised Pre-Training of Deep Learning Models with Contrastive Learning Improves Downstream Classification Performance

Self-supervised pre-training of deep learning models with contrastive learning is a widely used technique in image analysis. Current findings indicate a strong potential for contrastive pre-training on medical images. However, further research is necessary to incorporate the particular characteristics of these images. We hypothesize that the similarity of medical images hinders the success of contrastive learning in the medical imaging domain. To this end, we investigate different strategies based on deep embedding, information theory, and hashing in order to identify and reduce redundancy in medical pre-training datasets. The effect of these different reduction strategies on contrastive learning is evaluated on two pre-training datasets and several downstream classification tasks. In all of our experiments, dataset reduction leads to a considerable performance gain in downstream tasks, e.g., an AUC score improvement from 0.78 to 0.83 for the COVID CT Classification Grand Challenge, 0.97 to 0.98 for the OrganSMNIST Classification Challenge and 0.73 to 0.83 for a brain hemorrhage classification task. Furthermore, pre-training is up to nine times faster due to the dataset reduction. In conclusion, the proposed approach highlights the importance of dataset quality and provides a transferable approach to improve contrastive pre-training for classification downstream tasks on medical images.

eess.IV↗

arXiv2024

Geometry Dependent Localization of Surface Plasmons on Random Gold Nanoparticle Assemblies

Assemblies of plasmonic nanoparticles (NPs) support hybridized modes of localized surface plasmons (LSPs), which delocalize in geometrically well-ordered arrangements. Here, the hybridization behavior of LSPs in geometrically completely disordered arrangements of Au NPs fabricated by an e-beam synthesis method is studied. Employing electron energy loss spectroscopy in a scanning transmission electron microscope in combination with numerical simulations, the disorder-driven spatial and spectral localization of the coupled LSP modes that depend on the NP thickness is revealed. Below 0.4 nm sample thickness (flat NPs), localization increases towards higher hybridized LSP mode energies. In comparison, above 10 nm thickness, a decrease of localization (an increase of delocalization) with higher mode energies is observed. In the intermediate thickness regime, a transition of the energy dependence of the localization between the two limiting cases, exhibiting a transition mode energy with minimal localization, is observed. This behavior is mainly driven by the energy and thickness dependence of the polarizability of the individual NPs.

physics.optics↗

arXiv2024

Multidimensional asymptotic classes

We develop a general framework (multidimensional asymptotic classes, or m.a.c.s) for handling classes of finite first order structures with a strong uniformity condition on cardinalities of definable sets: The condition asserts that definable families given by a formula \phi(x,y) should take on a fixed number n_\phi of approximate sizes in any M in the class, with those sizes varying with M. The prototype is the class of all finite fields, where the uniformity is given by a theorem of Chatzidakis, van den Dries and Macintyre. It inspired the development of asymptotic classes of finite structures, which this new framework extends. The underlying theory of m.a.c.s is developed, including preservation under bi-interpretability, and a proof that for the m.a.c. condition to hold it suffices to consider formulas \phi(x,y) with x a single variable. Many examples of m.a.c.s are given, including 2-sorted structures (F,V) where V is a vector space over a finite field F possibly equipped with a bilinear form, and an example arising from representations of quivers of finite representation type. We also give examples and structural results for multidimensional exact classes (m.e.c.s), where the definable sets take a fixed number of precisely specified cardinalities, which again vary with M. We also develop a notion of infinite generalised measurable structure, whereby definable sets are assigned values in an ordered semiring. We show that any infinite ultraproduct of a m.a.c. is generalised measurable, that values can be taken in an ordered ring if the m.a.c. is a m.e.c., and explore model-theoretic consequences of generalised measurability. Such a structure cannot have the strict order property, and stability-theoretic properties can be read off from the measures in the semiring.

math.LO↗

arXiv2023

2D high temperature superconductor integration in contact printed circuit boards

Inherent properties of superconducting Bi2Sr2CaCu2O8+x films, such as the high superconducting transition temperature Tc, efficient Josephson coupling between neighboring CuO layers, and fast quasiparticle relaxation dynamics, make them a promising platform for advances in quantum computing and communication technologies. However, preserving two-dimensional superconductivity during device fabrication is an outstanding experimental challenge because of the fast degradation of the superconducting properties of two-dimensional flakes when they are exposed to moisture, organic sol vents, and heat. Here, to realize superconducting devices utilizing two-dimensional superconducting films, we develop a novel fabrication technique relying on the cryogenic dry transfer of printable circuits embedded into a silicon nitride membrane. This approach separates the circuit fabrication stage requiring chemically reactive substances and ionizing physical processes from the creation of the thin superconducting structures. Apart from providing electrical contacts in a single transfer step, the membrane encapsulates the surface of the crystal shielding it from the environment. The fabricated atomically thin Bi2Sr2CaCu2O8+x-based devices show high superconducting transition temperature Tc ~ 91 K close to that of the bulk crystal and demonstrate stable super conducting properties

cond-mat.supr-con↗

arXiv2023

Self-Supervised Pre-Training with Contrastive and Masked Autoencoder Methods for Dealing with Small Datasets in Deep Learning for Medical Imaging

Deep learning in medical imaging has the potential to minimize the risk of diagnostic errors, reduce radiologist workload, and accelerate diagnosis. Training such deep learning models requires large and accurate datasets, with annotations for all training samples. However, in the medical imaging domain, annotated datasets for specific tasks are often small due to the high complexity of annotations, limited access, or the rarity of diseases. To address this challenge, deep learning models can be pre-trained on large image datasets without annotations using methods from the field of self-supervised learning. After pre-training, small annotated datasets are sufficient to fine-tune the models for a specific task. The most popular self-supervised pre-training approaches in medical imaging are based on contrastive learning. However, recent studies in natural image processing indicate a strong potential for masked autoencoder approaches. Our work compares state-of-the-art contrastive learning methods with the recently introduced masked autoencoder approach "SparK" for convolutional neural networks (CNNs) on medical images. Therefore we pre-train on a large unannotated CT image dataset and fine-tune on several CT classification tasks. Due to the challenge of obtaining sufficient annotated training data in medical imaging, it is of particular interest to evaluate how the self-supervised pre-training methods perform when fine-tuning on small datasets. By experimenting with gradually reducing the training dataset size for fine-tuning, we find that the reduction has different effects depending on the type of pre-training chosen. The SparK pre-training method is more robust to the training dataset size than the contrastive methods. Based on our results, we propose the SparK pre-training for medical imaging tasks with only small annotated datasets.

cs.CV↗

arXiv2023

Twisted cuprate van der Waals heterostructures with controlled Josephson coupling

Twisted van der Waals (vdW) heterostructures offer a unique platform for engineering the efficient Josephson coupling between cuprate thin crystals harboring the nodal superconducting order parameter. Preparing vdW heterostructures-based Josephson junction comprising stacked cuprates requires maintaining an ordered interface with preserved surface superconductivity. Here, we report the preparation of the Josephson junction out of the stacked Bi2Sr2CuCa2O8+d crystals using the cryogenic dry transfer technique and encapsulating the junction with an insulating layer, that protects the interface during the electrical contacts evaporation at the 10-6 mbar base pressure. We find that the Josephson critical current Ic has a maximum at low twist angles, comparable to that of the bulk intrinsic Josephson junctions, and is reduced by two orders of magnitude at twist angles close to 45{\deg}. The reduction of Ic occurs due to a mismatch between superconducting d-wave order parameters, which suppresses the direct Cooper pair tunneling.

cond-mat.supr-con↗

arXiv2022

Role of substrate clamping on anisotropy and domain structure in the canted antiferromagnet $\alpha$-Fe$_2$O$_3$

Antiferromagnets have recently been propelled to the forefront of spintronics by their high potential for revolutionizing memory technologies. For this, understanding the formation and driving mechanisms of the domain structure is paramount. In this work, we investigate the domain structure in a thin-film canted antiferromagnet $\alpha$-Fe$_2$O$_3$. We find that the internal destressing fields driving the formation of domains do not follow the crystal symmetry of $\alpha$-Fe$_2$O$_3$, but fluctuate due to substrate clamping. This leads to an overall isotropic distribution of the N\'eel order with locally varying effective anisotropy in antiferromagnetic thin films. Furthermore, we show that the weak ferromagnetic nature of $\alpha$-Fe$_2$O$_3$ leads to a qualitatively different dependence on magnetic field compared to collinear antiferromagnets such as NiO. The insights gained from our work serve as a foundation for further studies of electrical and optical manipulation of the domain structure of antiferromagnetic thin films.

cond-mat.mes-hall↗

arXiv2022

Linear colossal magnetoresistance driven by magnetic textures in LaTiO3 thin films on SrTiO3

Linear magnetoresistance (LMR) is of particular interest for memory, electronics, and sensing applications, especially when it does not saturate over a wide range of magnetic fields. One of its principal origins is local mobility or density inhomogeneities, often structural, which in the Parish-Littlewood theory leads to an unsaturating LMR proportional to mobility. Structural disorder, however, also tends to limit the mobility and hence the overall LMR amplitude. An alternative route to achieve large LMR is via non-structural inhomogeneities which do not affect the zero field mobility, like magnetic domains. Here, linear positive magnetoresistance caused by magnetic texture is reported in \ch{LaTiO3}/\ch{SrTiO3} heterostructures. The LMR amplitude reaches up to 6500\% at 9T. This colossal value is understood by the unusual combination of a very high thin film mobility, up to 40 000 cm$^2$/V.s, and a very large coverage of low-mobility regions. These regions correlate with a striped magnetic structure, compatible with a spiral magnetic texture in the \ch{LaTiO3} film, revealed by low temperature Lorentz transmission electron microscopy. These results provide a novel route for the engineering of large-LMR devices.

cond-mat.mes-hall↗

arXiv2022

Electron Holographic Mapping of Structural and Electronic Reconstruction at Mono- and Bilayer Steps of h-BN

Here, by making use of medium and high resolution autocorrected off-axis electron holography, we directly probe the electrostatic potential as well as in-plane and out-of-plane charge delocalization at edges and steps in multilayer hexagonal boron nitride. In combination with ab-initio calculations, the data allows to directly reveal the formation of out-of-plane covalent bonds at folded zig-zag edges and steps comprising two monolayers and the absence of which at monolayer steps. The technique paves the way for studying other charge (de)localization phenomena in 2D materials, e.g., at polar edges, topological edge states and defects.

cond-mat.mtrl-sci↗

arXiv2022

Encapsulating high-temperature superconducting twisted van der Waals heterostructures blocks detrimental effects of disorder

High-temperature cuprate superconductors-based van der Waals (vdW) heterostructures hold high technological promise. One of the obstacles hindering progress is the detrimental effect of disorder on the properties of Josephson junctions (JJ) realized by the vdW devices. Here we report the new method of fabricating twisted vdW heterostructures made of Bi2Sr2CuCa2O8+d, crucially improving the JJ characteristics, pushing them up to those of the intrinsic JJs in bulk samples. The method combines a cryogenic stacking using a solvent-free stencil mask technique and covering interface by the insulating hexagonal boron nitride crystals. Despite the high-vacuum condition down to 10-6 mbar in the evaporation chamber, the interface appears to be protected from water molecules during the in-situ metal deposition only when fully encapsulated. Comparing the current-voltage curves of encapsulated and unencapsulated interfaces, we reveal that the encapsulated interfaces' characteristics are crucially improved so that the corresponding JJs demonstrate high critical currents and sharpness of the superconducting transition comparable to those of the intrinsic JJs. Finally, we show that the encapsulated heterostructures are more stable in time.

cond-mat.supr-con↗

arXiv2022

Optical and thermal analysis of the light-heat conversion process employing an antenna-based hybrid plasmonic waveguide for HAMR

We investigate a tapered, hybrid plasmonic waveguide which has previously been proposed as an optically efficient near-field transducer (NFT), or component thereof, in several devices which aim to exploit nanofocused light. We numerically analyze how light is transported through the waveguide and ultimately focused via effective-mode coupling and taper optimization. Crucial dimensional parameters in this optimization process are identified that are not only necessary to achieve maximum optical throughput, but also optimum thermal performance with specific application towards heat-assisted magnetic recording (HAMR). It is shown that existing devices constructed on similar waveguides may benefit from a heat spreader to avoid deformation of the plasmonic element which we achieve with no cost to the optical efficiency. For HAMR, our design is able to surpass many industry requirements in regard to both optical and thermal efficiency using pertinent figure of merits like 8.5% optical efficiency.

physics.optics↗

arXiv2021

Unveiling the three-dimensional spin texture of skyrmion tubes

Magnetic skyrmions are stable topological solitons with complex non-coplanar spin structures. Their nanoscopic size and the low electric currents required to initiate and control their motion has opened a new field of research, skyrmionics, that aims at using skyrmions as information carriers for data storage and manipulation. Recent advances in skyrmionics prompt for a thorough understanding of the detailed three-dimensional (3D) spin texture of a skyrmion including skyrmion-skyrmion interactions and their coupling to surfaces and interfaces. These properties crucially affect application-related aspects such as the stability and mobility of skyrmions in confined structures. To date, however, experimental techniques to measure the three-dimensional spin texture with nanometer resolution are largely missing. We therefore adapt holographic vector field electron tomography to the problem and report on the first quantitative reconstruction of the 3D spin texture of skyrmions with sub-10 nanometer resolution. The reconstructed textures reveal a variety of previously unseen local deviations from a homogeneous Bloch character within the skyrmion tubes (SkTs), details of the collapse of the skyrmion texture at surfaces, and a correlated modulation of the SkT in FeGe along their tube axes. The quantitative 3D data of the magnetic induction also allow to experimentally confirm some principles of skyrmion formation by deriving spatially resolved maps of the magnetic energy density across these magnetic solitons.

cond-mat.mtrl-sci↗

arXiv2020

Topological Hall effect arising from the mesoscopic and microscopic non-coplanar magnetic structure in MnBi

The topological Hall effect (THE), induced by the Berry curvature, which originates from non-zero scalar spin chirality, is an important feature for mesoscopic topological structures, such as skyrmions. However, the THE might also arise from other microscopic non-coplanar spin structures in the lattice. Thus, the origin of the THE inevitably needs to be determined to fully understand skyrmions and find new host materials. Here, we examine the Hall effect in both bulk- and micron-sized lamellar samples of MnBi. The sample size affects the temperature and field range in which the THE is detectable. Although bulk sample exhibits the THE only upon exposure to weak fields in the easy-cone state, in thin lamella the THE exists across a wide temperature range and occurs at fields near saturation. Our results show that both the non-coplanar spin structure in the lattice and topologically non-trivial skyrmion bubbles are responsible for the THE, and that the dominant mechanism depends on the sample size. Hence, the magnetic phase diagram for MnBi is strongly size-dependent. Our study provides an example in which the THE is simultaneously induced by two mechanisms, and builds a bridge between mesoscopic and microscopic magnetic structures.

cond-mat.mes-hall↗

arXiv2020

Autocorrected Off-axis Holography of 2D Materials

The reduced dimensionality in two-dimensional materials leads a wealth of unusual properties, which are currently explored for both fundamental and applied sciences. In order to study the crystal structure, edge states, the formation of defects and grain boundaries, or the impact of adsorbates, high resolution microscopy techniques are indispensible. Here we report on the development of an electron holography (EH) transmission electron microscopy (TEM) technique, which facilitates high spatial resolution by an automatic correction of geometric aberrations. Distinguished features of EH beyond conventional TEM imaging are the gap-free spatial information signal transfer and higher dose efficiency for certain spatial frequency bands as well as direct access to the projected electrostatic potential of the 2D material. We demonstrate these features at the example of h-BN, at which we measure the electrostatic potential as a function of layer number down to the monolayer limit and obtain evidence for a systematic increase of the potential at the zig-zag edges.

cond-mat.mes-hall↗