Ralf Meyer

Vibrational density of states of silicon nanoparticles

The vibrational density of states of silicon nanoparticles in the range from 2.3 to 10.3 nm is studied with the help of molecular-dynamics simulations. From these simulations the vibrational density of states and frequencies of bulk-like vibrational modes at high-symmetry points of the Brillouin-zone have been derived. The results show an increase of the density of states at low frequencies and a transfer of modes from the high-frequency end of the spectrum to the intermediate range. At the same time the peak of transverse optical modes is shifted to higher frequencies. These observations are in line with previous simulation studies of metallic nanoparticles and they provide an explanation for a previously observed discrepancy between experimental and theoretical data [C. Meier et al., Physica E, 32, 155 (2006)].

Numerical study of critical properties and hidden orders in dimerized spin ladders

Dimerized antiferromagnetic spin- 1 ladders are known to exhibit a quantum critical phase transition in the ground state, the existence or absence of which is dependent on the dimerization pattern of the ladder. The gapped phases cannot be distinguished by the conventional Landau long-range order parameter. However, they possess a non-local (hidden) string order parameter, which is nonzero in one phase and vanishes in the other. We use an exact diagonalization technique to calculate ground state energies, energy gaps and string order parameters of dimerized two- and three-leg Heisenberg ladders, as well as a critical scaling analysis to yield estimates of the critical exponents � and �.

Efficient parallelization of short-range molecular dynamics simulations on many-core systems.

This article introduces a highly parallel algorithm for molecular dynamics simulations with short-range forces on single node multi- and many-core systems. The algorithm is designed to achieve high parallel speedups for strongly inhomogeneous systems like nanodevices or nanostructured materials. In the proposed scheme the calculation of the forces and the generation of neighbor lists are divided into small tasks. The tasks are then executed by a thread pool according to a dependent task schedule. This schedule is constructed in such a way that a particle is never accessed by two threads at the same time. Benchmark simulations on a typical 12-core machine show that the described algorithm achieves excellent parallel efficiencies above 80% for different kinds of systems and all numbers of cores. For inhomogeneous systems the speedups are strongly superior to those obtained with spatial decomposition. Further benchmarks were performed on an Intel Xeon Phi coprocessor. These simulations demonstrate that the algorithm scales well to large numbers of cores.

A hybrid algorithm for parallel molecular dynamics simulations

Abstract This article describes algorithms for the hybrid parallelization and SIMD vectorization of molecular dynamics simulations with short-range forces. The parallelization method combines domain decomposition with a thread-based parallelization approach. The goal of the work is to enable efficient simulations of very large (tens of millions of atoms) and inhomogeneous systems on many-core processors with hundreds or thousands of cores and SIMD units with large vector sizes. In order to test the efficiency of the method, simulations of a variety of configurations with up to 74 million atoms have been performed. Results are shown that were obtained on multi-core systems with Sandy Bridge and Haswell processors as well as systems with Xeon Phi many-core processors.

Vibrational properties of metallic nanoparticles

This brief overview discusses the structure and vibrational dynamics of metallic nanoparticles. The theoretical results are derived from molecular dynamics simulations using empirical tight-binding potentials. The vibrational densities of states of nanoparticles with sizes from 4 to 20 nm are shown. A method to identity surface, subsurface and core atoms is presented and it is shown that the average vibrational density of states of the atoms in these regions is fairly independent of the particle size.

Simulating Structure, Magnetism and Electronic Properties of Monoatomic, Binary and Ternary Transition Metal Nanoclusters

Based on large‐scale molecular dynamics and density functional theory calculations, we provide some insight into morphological, magnetic and electronic changes of monoatomic (Fe), binary (Fe‐Pt) and ternary (Ni‐Mn‐Ga) transition metal clusters with the size of the clusters. For Fe, the magnetic order is shown to influence the cluster growth and to stabilize the bcc like structure already for small cluster sizes. For Fe‐Pt clusters up to a few nm in size, multiply‐twinned structures are more favorable than the L10 phase. For not too big ternary Ni‐Mn‐Ga clusters, the martensitic tendency of cubic to tetragonal transformation is like in Fe‐Pt hindered by surface effects, while the bigger clusters develop a bulk‐like tetragonal transformation.

The Martensitic Transformation in Iron-Nickel Alloys: A Molecular Dynamics Study

Many metals and metallic alloys show martensitic transformations at temperatures below the melting point. Martensitic transformations are structural phase changes of first order which belong to the broader class of diffusionless solid-state phase transformations. These are structural transformations of the crystal lattice, which do not involve long-range atomic movements. A recent review of the properties and the classification of diffusionless transformations has been given by Delaye1.

Vibrational band structure of nanoscale phononic crystals

The vibrational properties of two-dimensional phononic crystals are studied with large-scale molecular dynamics simulations and finite element method calculation. The vibrational band structure derived from the molecular dynamics simulations shows the existence of partial acoustic band gaps along the Γ–M direction. The band structure is in excellent agreement with the results from the finite element model, proving that molecular dynamics simulations can be used to study the vibrational properties of such complex systems. An analysis of the structure of the vibrational modes reveals how the acoustic modes deviate from the homogeneous bulk behavior for shorter wavelengths and hints toward a decoupling of vibrations in the phononic crystal.

Molecular-Dynamics Simulations Using Spatial Decomposition and Task-Based Parallelism

This article discusses the implementation of a hybrid algorithm for the parallelization of molecular-dynamics simulations. The hybrid algorithm combines the spatial decomposition method using message passing with a task-based, threaded approach for the parallelization of the workload. Benchmark simulations on a multi-core system and an Intel Xeon Phi co-processor show that the hybrid algorithm provides better performances than the message-passing or threaded approaches alone.

Application of advanced diagonalization methods to quantum spin systems.

In this work, the Block Davidson and the residual minimization-direct inversion in the iterative subspace (RMM-DIIS) algorithms are used to diagonalize the Hamilton matrices arising from antiferromagnetic spin-\(\frac{1}{2}\) Heisenberg models. The results show that both algorithms find reliably the lowest eigenvalues but the computational costs are smaller for the RMM-DIIS method. In addition to this, the authors show that the new Intel Xeon Phi coprocessor can be used efficiently for this type of problems.