Three parallel algorithms for classical molecular dynamics are presented. The first assigns each processor a fixed subset of atoms; the second assigns each a fixed subset of inter-atomic forces to compute; the third assigns each a fixed spatial region. The algorithms are suitable for molecular dynamics models which can be difficult to parallelize efficiently—those with short-range forces where the neighbors of each atom change rapidly. They can be implemented on any distributed-memory parallel machine which allows for message-passing of data between independently executing processors. The algorithms are tested on a standard Lennard-Jones benchmark problem for system sizes ranging from 500 to 100,000,000 atoms on several parallel supercomputers--the nCUBE 2, Intel iPSC/860 and Paragon, and Cray T3D. Comparing the results to the fastest reported vectorized Cray Y-MP and C90 algorithm shows that the current generation of parallel machines is competitive with conventional vector supercomputers even for small problems. For large problems, the spatial algorithm achieves parallel efficiencies of 90% and a 1840-node Intel Paragon performs up to 165 faster than a single Cray C9O processor. Trade-offs between the three algorithms and guidelines for adapting them to more complex molecular dynamics simulations are also discussed.

Fast parallel algorithms for short-range molecular dynamics

The tetragonal to orthorhombic ferroelastic phase transition between rutile- and ${\text{CaCl}}_{2}$-type ${\text{SiO}}_{2}$ at high pressures is studied using first-principles calculations and the Landau free-energy expansion. The phase transition is systematically investigated in terms of characteristic phonon modes with ${\text{B}}_{1g}$ and ${\text{A}}_{g}$ symmetries, shear moduli, transverse-acoustic mode, rotation angle of the ${\text{SiO}}_{6}$ octahedra, spontaneous symmetry-breaking and volume strains, and enthalpy. The results show that these physical behaviors at the transition are well described using the Landau free-energy expansion parametrized by the first-principles calculations.

/pdf/first-principles-calculations-of-the-ferroelastic-transition-4b3qx4ec16.pdf

First-principles calculations of the ferroelastic transition between rutile-type and CaCl2-type SiO2 at high pressures

During the past decade, computer simulations based on a quantum-mechanical description of the interactions between electrons and between electrons and atomic nuclei have developed an increasingly important impact on solid-state physics and chemistry and on materials science—promoting not only a deeper understanding, but also the possibility to contribute significantly to materials design for future technologies. This development is based on two important columns: (i) The improved description of electronic many-body effects within density-functional theory (DFT) and the upcoming post-DFT methods. (ii) The implementation of the new functionals and many-body techniques within highly efficient, stable, and versatile computer codes, which allow to exploit the potential of modern computer architectures. In this review, I discuss the implementation of various DFT functionals [local-density approximation (LDA), generalized gradient approximation (GGA), meta-GGA, hybrid functional mixing DFT, and exact (Hartree-Fock) exchange] and post-DFT approaches [DFT + U for strong electronic correlations in narrow bands, many-body perturbation theory (GW) for quasiparticle spectra, dynamical correlation effects via the adiabatic-connection fluctuation-dissipation theorem (AC-FDT)] in the Vienna ab initio simulation package VASP. VASP is a plane-wave all-electron code using the projector-augmented wave method to describe the electron-core interaction. The code uses fast iterative techniques for the diagonalization of the DFT Hamiltonian and allows to perform total-energy calculations and structural optimizations for systems with thousands of atoms and ab initio molecular dynamics simulations for ensembles with a few hundred atoms extending over several tens of ps. Applications in many different areas (structure and phase stability, mechanical and dynamical properties, liquids, glasses and quasicrystals, magnetism and magnetic nanostructures, semiconductors and insulators, surfaces, interfaces and thin films, chemical reactions, and catalysis) are reviewed. © 2008 Wiley Periodicals, Inc. J Comput Chem, 2008

Ab-initio simulations of materials using VASP: Density-functional theory and beyond.

We present a nonempirical density functional generalized gradient approximation (GGA) that gives significant improvements for lattice constants, crystal structures, and metal surface energies over the most popular Perdew-Burke-Ernzerhof (PBE) GGA. The functional is based on a diffuse radial cutoff for the exchange hole in real space, and the analytic gradient expansion of the exchange energy for small gradients. There are no adjustable parameters, the constraining conditions of PBE are maintained, and the functional is easily implemented in existing codes.

More accurate generalized gradient approximation for solids

https://www.perplex.ethz.ch/papers/connolly_epsl_05.pdf

Computation of phase equilibria by linear programming: A tool for geodynamic modeling and its application to subduction zone decarbonation

P>We complete the development and description of a thermodynamic method for the computation of phase equilibria and physical properties of multiphase mantle assemblages. Our previous paper focused on the computation of physical properties. In this paper, our focus shifts to the phase equilibria. We further develop our theory to specify the ideal and excess contributions to solution properties and derive properties of multiphase assemblages. We discuss our global inversion strategy for determining the values of the free parameters in our theory and compare inverted parameter values with expectations based on scaling arguments. Comparisons between our method and experimental phase equilibria data encompass the pressure-temperature regime of Earth's mantle. Finally, we present applications of our method that illustrate how it may be used to explore the origins of mantle structure and mantle dynamics. Continuing rapid advances in experimental and theoretical petrology and mineral physics have motivated an expansion of the scope of our model via the addition of several new phases, and of the soda component: an appendix lists all parameters in our model and references to the experimental and theoretical studies that constrain them. Our algorithm for global minimization of the Gibbs free energy is embodied in a code called HeFESTo, and is detailed in a second appendix.

/pdf/thermodynamics-of-mantle-minerals-ii-phase-equilibria-151bu7omt0.pdf

Thermodynamics of mantle minerals - II. Phase equilibria

SUMMARY We present a theory for the computation of phase equilibria and physical properties of multicomponent assemblages relevant to the mantle of the Earth. The theory differs from previous treatments in being thermodynamically self-consistent: the theory is based on the concept of fundamental thermodynamic relations appropriately generalized to anisotropic strain and in encompassing elasticity in addition to the usual isotropic thermodynamic properties. In this first paper, we present the development of the theory, discuss its scope, and focus on its application to physical properties of mantle phases at elevated pressure and temperature including the equation of state, thermochemical properties and the elastic wave velocities. We find that the Eulerian finite strain formulation captures the variation of the elastic moduli with compression. The variation of the vibrational frequencies with compression is also cast as a Taylor series expansion in the Eulerian finite strain, the appropriate volume derivative of which leads �

/pdf/thermodynamics-of-mantle-minerals-i-physical-properties-1qtvvk058q.pdf

Thermodynamics of mantle minerals – I. Physical properties

We compare the predictions of compositional models of the mantle transition zone to observed seismic properties by constructing phase diagrams in the MgO-FeO-CaO-Al2O3-SiO2 system and estimating the elasticity of the relevant minerals. Mie-Gruneisen and Birch-Murnaghan finite strain theory are combined with ideal solution theory to extrapolate experimental measurements of thermal and elastic properties to high pressures and temperatures. The resulting thermodynamic potentials are combined with the estimated phase diagrams to predict the density, seismic parameter, and mantle adiabats for a given compositional model. We find that the properties of pyrolite agree well with the observed density and bulk sound velocity of the upper mantle and transition zone. Piclogite significantly underestimates the magnitude of the 400-km velocity discontinuity and overestimates the velocity gradient in the transition zone. Substantially enriching piclogite in Al provides an acceptable fit to the observations. Invoking a chemical boundary layer between the uppermost mantle and transition zone leads to poor agreement with observed seismic properties for the compositions considered. Within the transition zone, the dissolution of garnet to Ca-perovskite near 18 GPa may explain the proposed 520-km seismic discontinuity. Below 700 km depth, all compositions disagree with observed bulk sound velocities, implying that the lower mantle is chemically distinct from the upper mantle.

Petrology, elasticity, and composition of the mantle transition zone

The structural and elastic properties of MgO periclase were studied up to 150 GPa with the first-principles pseudopotential method within the local density approximation. The calculated lattice constant of the B1 phase over the pressure range studied is within 1% of experimental values. The observed B1 phase of MgO was found to be stable up to 450 GPa, precluding the B1-B2 phase transition within the lower mantle. The calculated transition pressure is less than one-half of the previous pseudopotential prediction but is very close to the linearized augmented plane-wave result. All three independent elastic constants, c(11), c(12), and c(44) for the B1 phase are calculated from direct computation of stresses generated by small strains. The calculated zero-pressure values of the elastic moduli and wave velocities and their initial pressure dependence are in excellent agreement with experiments. MgO was found to be highly anisotropic in its elastic properties, with the magnitude of the anisotropy first decreasing between 0 and 15 Gpa and then increasing from 15 to 150 GPa. Longitudinal and shear-wave velocities were found to vary by 23 and 59%, respectively, with propagation direction at 150 Gpa. The character of the anisotropy changes qualitatively with pressure. At zero pressure longitudinal and shear-wave propagations are fastest along [111] and [100], respectively, whereas above 15 GPa, the corresponding fast directions are [100] and [110]. The Cauchy condition was found to be strongly violated in MgO, reflecting the importance of noncentral many-body forces.

https://www.degruyter.com/downloadpdf/j/ammin.1997.82.issue-1-2/am-1997-1-207/am-1997-1-207.xml

Lars Stixrude

Papers

Thermodynamics of mantle minerals - II. Phase equilibria

Thermodynamics of mantle minerals – I. Physical properties

Petrology, elasticity, and composition of the mantle transition zone

Structure and elasticity of MgO at high pressure

Earth's Deep Interior: Mineral Physics and Tomography From the Atomic to the Global Scale