Search

Computational science is crucial for delivering reliable weather and climate predictions. However, despite decades of high-performance computing experience, there is serious concern about the sustainability of this application in the post-Moore/Dennard era. Here, we discuss the present limitations in the field and propose the design of a novel infrastructure that is scalable and more adaptable to future, yet unknown computing architectures., (© 2021. Springer Nature America, Inc.)

Materials Cloud is a platform designed to enable open and seamless sharing of resources for computational science, driven by applications in materials modelling. It hosts (1) archival and dissemination services for raw and curated data, together with their provenance graph, (2) modelling services and virtual machines, (3) tools for data analytics, and pre-/post-processing, and (4) educational materials. Data is citable and archived persistently, providing a comprehensive embodiment of entire simulation pipelines (calculations performed, codes used, data generated) in the form of graphs that allow retracing and reproducing any computed result. When an AiiDA database is shared on Materials Cloud, peers can browse the interconnected record of simulations, download individual files or the full database, and start their research from the results of the original authors. The infrastructure is agnostic to the specific simulation codes used and can support diverse applications in computational science that transcend its initial materials domain.

We have investigated the group 14 nitrides (M3N4) in the spinel phase (γ-M3N4 with M = C, Si, Ge, and Sn) and β phase (β-M3N4 with M = Si, Ge, and Sn) using density functional theory with the local density approximation and the GW approximation. The Kohn-Sham energies of these systems have been first calculated within the framework of full-potential linearized augmented plane waves (LAPW) and then corrected using single-shot G0W0 calculations, which we have implemented in the modified version of the Elk full-potential LAPW code. Direct band gaps at the Γ point have been found for spinel-type nitrides γ-M3N4 with M = Si, Ge, and Sn. The corresponding GW-corrected band gaps agree with experiment. We have also found that the GW calculations with and without the plasmon-pole approximation give very similar results, even when the system contains semi-core d electrons. These spinel-type nitrides are novel materials for potential optoelectronic applications because of their direct and tunable band gaps.

The density of states g (ε) is frequently used to calculate the temperature-dependent properties of a thermodynamic system. Here a derivation is given for calculating the warped density of states g*(ε) resulting from the addition of a perturbation. The method is validated for a classical Heisenberg model of bcc Fe and the errors in the free energy are shown to be second order in the perturbation. Taking the perturbation to be the difference between a first-principles quantum-mechanical energy and a corresponding classical energy, this method can significantly reduce the computational effort required to calculate g(ε) for quantum systems using the Wang-Landau approach.

Using dynamic cluster quantum Monte Carlo simulations, we study the superconducting behavior of a 1/8 doped two-dimensional Hubbard model with imposed unidirectional stripelike charge-density-wave modulation. We find a significant increase of the pairing correlations and critical temperature relative to the homogeneous system when the modulation length scale is sufficiently large. With a separable form of the irreducible particle-particle vertex, we show that optimized superconductivity is obtained for a moderate modulation strength due to a delicate balance between the modulation enhanced pairing interaction, and a concomitant suppression of the bare particle-particle excitations by a modulation reduction of the quasiparticle weight.

In studying well-characterized, exchange-biased Fe(3)O(4)/CoO superlattices, we demonstrate a causal link between the exchange bias effect and the perpendicular coupling of the ferrimagnetic and antiferromagnetic spins. Neutron diffraction studies reveal that for thin CoO layers the onset temperature for exchange biasing T(B) matches the onset of locked-in, preferential perpendicular coupling of the spins, rather than the antiferromagnetic ordering temperature T(N). The results are explained by considering the role of anisotropic exchange first proposed by Dzyaloshinsky and Moriya and developing a model based purely on information on structural defects and exchange for these oxides. The devised mechanism provides a general explanation of biasing in systems with perpendicular coupling.

We present a modified Wang-Landau algorithm for models with continuous degrees of freedom. We demonstrate this algorithm with the calculation of the joint density of states of ferromagnet Heisenberg models and a model polymer chain. The joint density of states contains more information than the density of states of a single variable-energy, but is also much more time consuming to calculate. We present strategies to significantly speed up this calculation for large systems over a large range of energy and order parameter.

The cluster size dependence of superconductivity in the conventional two-dimensional Hubbard model, commonly believed to describe high-temperature superconductors, is systematically studied using the dynamical cluster approximation and quantum Monte Carlo simulations as a cluster solver. Because of the nonlocality of the d-wave superconducting order parameter, the results on small clusters show large size and geometry effects. In large enough clusters, the results are independent of the cluster size and display a finite temperature instability to d-wave superconductivity.

Large scale computer simulations are used to elucidate a long-standing controversy regarding the existence, or otherwise, of spin waves in paramagnetic bcc iron. Spin dynamics simulations of the dynamic structure factor of a Heisenberg model of Fe with first principles interactions reveal that well defined peaks persist far above Curie temperature Tc. At large wave vectors these peaks can be ascribed to propagating spin waves; at small wave vectors the peaks correspond to overdamped spin waves. Paradoxically, spin wave excitations exist despite only limited magnetic short-range order at and above Tc.

Identifying the specific DNA-binding sites of transcription-factor proteins is essential to understanding the regulation of gene expression in the cell. Bioinformatics approaches are fast compared to experiments, but require prior knowledge of multiple binding sites for each protein. Here, we present an atomistic force-field method to predict binding sites based only on the X-ray structure of a related bound complex. Specific flexible contacts between the protein and DNA are modeled by a library of amino acid side-chain rotamers. Using the example of the mouse transcription factor, Zif268, a well-studied zinc-finger protein, we show that the protein sequence alone, without the detailed experimental structure, gives a strong bias toward the consensus binding site., (Copyright 2004 Wiley-Liss, Inc.)

We report collective ferromagnetic behavior with high Curie temperatures (T(c)) in Fe dot assemblies supported by the Cu(111) surface. Our ability to tune the average size and spacing of the individual dots allows us to conclude that enhanced magnetic anisotropy cannot account for this high-T(c) ferromagnetic order. Because our Monte Carlo simulations have ruled out the dipolar interaction as the dominant factor in this system, we attribute the origin of the ferromagnetic order to indirect exchange coupling via the Cu(111) substrate.