arXiv subjects

Florian Günther

Publications and source records attributed to Florian Günther.

11 recordsLinked to original sources

arXiv2026

A Symmetry-Constrained Fourier--Morse Framework for Compact Anisotropic Interaction Potentials

Large-scale coarse-grained simulations of anisotropic particles require compact interaction models that retain orientation-dependent energetics. We present a symmetry-constrained Fourier--Morse framework in which the radial interaction is described by a Morse potential and its orientational dependence by Fourier expansions. The representation converges systematically with harmonic resolution, allows known orientational symmetries to be imposed directly, and supports further reduction through harmonic truncation and coefficient pruning. Its explicit Fourier structure also provides a natural basis for constructing or modifying model interactions with prescribed orientational symmetries. The parameterization requires only a sampled interaction landscape and is therefore independent of the method used to generate the reference data. We demonstrate the approach for four interaction classes of chiral $α$-polyalanine helices, representing more than \num{300000} reference energy values with tens to a few hundred coefficients while reproducing equilibrium interaction features with meV- and mÅ-level errors. As a proof of concept, molecular-dynamics simulations using the reduced analytical potentials produce stable low-temperature configurations exhibiting local ordering motifs qualitatively consistent with those identified previously by Monte Carlo simulated annealing.

physics.comp-ph↗

arXiv2026

Current-Gated Nonlinear Dynamics of a Self-Heating Memristor: an Electrothermal Extension of the Pickett Filamentary Model

Self-heating couples the electrical and thermal states of filamentary memristors. However, the widely used Pickett compact model of $TiO_2$ resistive switching is isothermal and therefore cannot capture the resulting electrothermal dynamics. We introduce the Arrhenius-Thermal Filamentary Model (ATFM), which extends Pickett's tunneling-gap kinetics by incorporating a dynamic heat balance and an Arrhenius-activated switching rate. The resulting electrothermal feedback produces a sharp current-gated transition: below a critical drive current, the tunneling gap undergoes a non-returning ratchet drift, whereas above it, exponential locking of the filament kinetics establishes a bounded, drive-locked electrothermal oscillation. Using a stroboscopic Poincaré map and the Floquet multipliers of the resulting period-$1$ orbit, we characterize this onset as a threshold-like orbit contraction rather than a classical local bifurcation. In the limit $E_a\to0$, ATFM recovers the isothermal Pickett dynamics to numerical precision, as verified against an independent reference implementation over amplitude, frequency, and activation-energy sweeps. A variance-based Sobol' analysis with bootstrap confidence intervals identifies the excitation amplitude as the dominant control parameter and the thermal resistance $R_{th}$, rather than the thermal capacitance $C_{th}$, as the leading thermal contributor. A geometry-dependent temperature constraint further reveals a non-monotonic operating window in which an intermediate active area maximizes the switching excursion. The predicted trajectories are reproduced by both a fully behavioral SPICE netlist and a Verilog-A/OSDI device implementation, making ATFM directly suitable for circuit simulation. Overall, ATFM reveals and realizes a self-heating-driven dynamical regime within the widely used Pickett filamentary framework.

nlin.CD↗

arXiv2026

Simulation of Self-Assembled Monolayers of Polyalanine $α$-Helices: Development and Application of an Effective Potential for Film Structure Predictions

Self-assembled monolayers of polyalanine $α$-helices exhibit distinct structural phases with implications for chiral-induced spin selectivity. We combine scanning tunneling microscopy and theoretical modeling to reveal how chiral composition governs supramolecular organization. Enantiopure systems form hexagonal lattices, while racemic mixtures organize into rectangular phases with stripe-like features. Our interaction potentials derived from density-functional based tight binding calculations show that opposite-handed helix pairs exhibit stronger binding and closer packing, explaining the denser racemic structures. Crucially, we demonstrate that the observed STM contrast arises from anti-parallel alignment of opposite-handed helices rather than physical height variations. These findings establish fundamental structure-property relationships for designing peptide-based spintronic materials.

physics.comp-ph↗

arXiv2025

IDEAL-M3D: Instance Diversity-Enriched Active Learning for Monocular 3D Detection

Monocular 3D detection relies on just a single camera and is therefore easy to deploy. Yet, achieving reliable 3D understanding from monocular images requires substantial annotation, and 3D labels are especially costly. To maximize performance under constrained labeling budgets, it is essential to prioritize annotating samples expected to deliver the largest performance gains. This prioritization is the focus of active learning. Curiously, we observed two significant limitations in active learning algorithms for 3D monocular object detection. First, previous approaches select entire images, which is inefficient, as non-informative instances contained in the same image also need to be labeled. Secondly, existing methods rely on uncertainty-based selection, which in monocular 3D object detection creates a bias toward depth ambiguity. Consequently, distant objects are selected, while nearby objects are overlooked. To address these limitations, we propose IDEAL-M3D, the first instance-level pipeline for monocular 3D detection. For the first time, we demonstrate that an explicitly diverse, fast-to-train ensemble improves diversity-driven active learning for monocular 3D. We induce diversity with heterogeneous backbones and task-agnostic features, loss weight perturbation, and time-dependent bagging. IDEAL-M3D shows superior performance and significant resource savings: with just 60% of the annotations, we achieve similar or better AP3D on KITTI validation and test set results compared to training the same detector on the whole dataset.

cs.CV↗

arXiv2020

Equilibrium orientation and adsorption of an ellipsoidal Janus particle at a fluid-fluid interface

We investigate the equilibrium orientation and adsorption process of a single, ellipsoidal Janus particle at a fluid-fluid interface. The particle surface comprises equally sized parts that are hydrophobic or hydrophilic. We present free energy models to predict the equilibrium orientation and compare the theoretical predictions with lattice Boltzmann simulations. We find that the deformation of the fluid interface strongly influences the equilibrium orientation of the Janus ellipsoid. The adsorption process of the Janus ellipsoid can lead to different final orientations determined by the interplay of particle aspect ratio and particle wettablity contrast.

cond-mat.soft↗

arXiv2015

Tunable Dipolar Capillary Deformations for Magnetic Janus Particles at Fluid-Fluid Interfaces

Janus particles have attracted significant interest as building blocks for complex materials in recent years. Furthermore, capillary interactions have been identified as a promising tool for directed self-assembly of particles at fluid-fluid interfaces. In this paper, we develop theoretical models describing the behaviour of magnetic Janus particles adsorbed at fluid-fluid interfaces interacting with an external magnetic field. Using numerical simulations, we test the models predictions and show that the magnetic Janus particles deform the interface in a dipolar manner. We suggest how to utilise the resulting dipolar capillary interactions to assemble particles at a fluid-fluid interface, and further demonstrate that the strength of these interactions can be tuned by altering the external field strength, opening up the possibility to create novel, reconfigurable materials.

cond-mat.soft↗

arXiv2014

Domain and droplet sizes in emulsions stabilized by colloidal particles

Particle-stabilized emulsions are commonly used in various industrial applications. These emulsions can present in different forms, such as Pickering emulsions or bijels, which can be distinguished by their different topologies and rheology. We numerically investigate the effect of the volume fraction and the uniform wettability of the stabilizing spherical particles in mixtures of two fluids. For this, we use the well-established three-dimensional lattice Boltzmann method, extended to allow for the added colloidal particles with non-neutral wetting properties. We obtain data on the domain sizes in the emulsions by using both structure functions and the Hoshen-Kopelman (HK) algorithm, and demonstrate that both methods have their own (dis-)advantages. We confirm an inverse dependence between the concentration of particles and the average radius of the stabilized droplets. Furthermore, we demonstrate the effect of particles detaching from interfaces on the emulsion properties and domain size measurements.

cond-mat.soft↗

arXiv2014

Timescales of emulsion formation caused by anisotropic particles

Particle stabilized emulsions have received an enormous interest in the recent past, but our understanding of the dynamics of emulsion formation is still limited. For simple spherical particles, the time dependent growth of fluid domains is dominated by the formation of droplets, particle adsorption and coalescence of droplets (Ostwald ripening), which eventually can be almost fully blocked due to the presence of the particles. Ellipsoidal particles are known to be more efficient stabilizers of fluid interfaces than spherical particles and their anisotropic shape and the related additional rotational degrees of freedom have an impact on the dynamics of emulsion formation. In this paper, we investigate this point by means of simple model systems consisting of a single ellipsoidal particle or a particle ensemble at a flat interface as well as a particle ensemble at a spherical interface. By applying combined multicomponent lattice Boltzmann and molecular dynamics simulations we demonstrate that the anisotropic shape of ellipsoidal particles causes two additional timescales to be of relevance in the dynamics of emulsion formation: a relatively short timescale can be attributed to the adsorption of single particles and the involved rotation of particles towards the interface. As soon as the interface is jammed, however, capillary interactions between the particles cause a local reordering on very long timescales leading to a continuous change in the interface configuration and increase of interfacial area. This effect can be utilized to counteract the thermodynamic instability of particle stabilized emulsions and thus offers the possibility to produce emulsions with exceptional stability.

cond-mat.soft↗

arXiv2013

Numerical simulations of complex fluid-fluid interface dynamics

Interfaces between two fluids are ubiquitous and of special importance for industrial applications, e.g., stabilisation of emulsions. The dynamics of fluid-fluid interfaces is difficult to study because these interfaces are usually deformable and their shapes are not known a priori. Since experiments do not provide access to all observables of interest, computer simulations pose attractive alternatives to gain insight into the physics of interfaces. In the present article, we restrict ourselves to systems with dimensions comparable to the lateral interface extensions. We provide a critical discussion of three numerical schemes coupled to the lattice Boltzmann method as a solver for the hydrodynamics of the problem: (a) the immersed boundary method for the simulation of vesicles and capsules, the Shan-Chen pseudopotential approach for multi-component fluids in combination with (b) an additional advection-diffusion component for surfactant modelling and (c) a molecular dynamics algorithm for the simulation of nanoparticles acting as emulsifiers.

cond-mat.soft↗

arXiv2012

Effects of nanoparticles and surfactant on droplets in shear flow

We present three-dimensional numerical simulations, employing the well-established lattice Boltzmann method, and investigate similarities and differences between surfactants and nanoparticles as additives at a fluid-fluid interface. We report on their respective effects on the surface tension of such an interface. Next, we subject a fluid droplet to shear and explore the deformation properties of the droplet, its inclination angle relative to the shear flow, the dynamics of the particles at the interface, and the possibility of breakup. Particles are seen not to affect the surface tension of the interface, although they do change the overall interfacial free energy. The particles do not remain homogeneously distributed over the interface, but form clusters in preferred regions that are stable for as long as the shear is applied. However, although the overall structure remains stable, individual nanoparticles roam the droplet interface, with a frequency of revolution that is highest in the middle of the droplet interface, normal to the shear flow, and increases with capillary number. We recover Taylor's law for small deformation of droplets when surfactant or particles are added to the droplet interface. The effect of surfactant is captured in the capillary number, but the inertia of adsorbed massive particles increases deformation at higher capillary number and eventually leads to easier breakup of the droplet.

cond-mat.soft↗

arXiv2012

Lattice Boltzmann simulations of anisotropic particles at liquid interfaces

Complex colloidal fluids, such as emulsions stabilized by complex shaped particles, play an important role in many industrial applications. However, understanding their physics requires a study at sufficiently large length scales while still resolving the microscopic structure of a large number of particles and of the local hydrodynamics. Due to its high degree of locality, the lattice Boltzmann method, when combined with a molecular dynamics solver and parallelized on modern supercomputers, provides a tool that allows such studies. Still, running simulations on hundreds of thousands of cores is not trivial. We report on our practical experiences when employing large fractions of an IBM Blue Gene/P system for our simulations. Then, we extend our model for spherical particles in multicomponent flows to anisotropic ellipsoidal objects rendering the shape of e.g. clay particles. The model is applied to a number of test cases including the adsorption of single particles at fluid interfaces and the formation and stabilization of Pickering emulsions or bijels.

cond-mat.soft↗