We present results of an ab-initio study of the electronic structure of 140 rare earth compounds. Specifically we predict an electronic phase diagram of the entire range of rare earth monopnictides and monochalcogenides, composed of metallic, semiconducting and heavy fermion-like regions, and exhibiting valency transitions brought about by a complex interplay between ligand chemistry and lanthanide contraction. The calculations exploit the combined effect of a first-principles methodology, which can adequately describe the dual character of electrons, itinerant vs. localized, and high throughput computing made possible by the increasing available computational power. Our findings, including the predicted "intermediate valent" compounds SmO and TmSe, are in overall excellent agreement with the available experimental data. The accuracy of the approach, proven e.g. through the lattice parameters calculated to within ∼1.5% of the experimental values, and its ability to describe localization phenomena in solids, makes it a competitive atomistic simulation approach in the search for and design of new materials with specific physical properties and possible technological applications. PACS numbers:

Results of ab initio calculations using the relativistic Local Spin Density theory are presented for the magnetic moments of periodic 5d and 4d transition metal interfaces with bcc Fe(001). In this systematic study we calculated the layer-resolved spin and orbital magnetic moments over the entire series. For the Fe/W(001) system, the Fe spin moment is reduced whilst its orbital moment is strongly enhanced. In the W layers a spin moment is induced, which is antiparallel to that of Fe in the first and fourth W layers but parallel to Fe in the second and third W layers. The W orbital moment does not follow the spin moment. It is aligned antiparallel to Fe in the first two W layers and changes sign in the third and fourth W layers. Therefore, Hund's third rule is violated in the first and third W layers, but not in the second and fourth W layers. The trend in the spin and orbital moments over the 4d and 5d series for multilayers is quite similar to previous impurity calculations. These observations strongly suggest that these effects can be seen as a consequence of the hybridization between 5d (4d) and Fe which is mostly due to band filling, and to a lesser extent geometrical effects of either single impurity or interface.

The first-principles SIC-LSD theory is utilized to study electronic, magnetic and orbital phenomena in LaMnO3. The correct ground state is found, which is antiferro orbitally ordered with the spin magnetic moments antiferromagnetically aligned. Jahn-Teller energies are found to be the largest energy scale. In addition it is the Jahn-Teller interaction which is the dominant effect in realising orbital order, and the electronic effects alone do not suffice.

We propose an explanation for the anomalous compressibility maximum in amorphous silica based on rigidity arguments. The model considers the fact that a network structure will be rigidly compressed in the high-pressure limit, and rigidly taut in the negative pressure limit, but flexible and hence softer at intermediate pressures. We validate the plausibility of this explanation by the analysis of molecular dynamics simulations. In fact this model is quite general, and will apply to any network solid, crystalline or amorphous; there are experimental indications that support this prediction. In contrast to other ideas concerning the compressibility maximum in amorphous silica, the model presented here does not invoke the existence of polyamorphic phase transitions in the glass phase.

This paper describes a prototype grid infrastructure, called the "eMinerals minigrid", for molecular simulation scientists. which is based on an integration of shared compute and data resources. We describe the key components, namely the use of Condor pools, Linux/Unix clusters with PBS and IBM's LoadLeveller job handling tools, the use of Globus for security handling, the use of Condor-G tools for wrapping globus job submit commands, Condor's DAGman tool for handling workflow, the Storage Resource Broker for handling data, and the CCLRC dataportal and associated tools for both archiving data with metadata and making data available to other workers.