arXiv subjects

Jose M. Soler

Publications and source records attributed to Jose M. Soler.

34 records · Page 2Linked to original sources

arXiv2006

Magnetism of two-dimensional defects in Pd: stacking faults, twin boundaries and surfaces

Careful first-principles density functional calculations reveal the importance of hexagonal versus cubic stacking of closed packed planes of Pd as far as local magnetic properties are concerned. We find that, contrary to the stable face centered cubic phase, which is paramagnetic, the hexagonal close-packed phase of Pd is ferromagnetic with a magnetic moment of 0.35 $μ_{B}$/atom. Our results show that two-dimensional defects with local hcp stacking, like twin boundaries and stacking faults, in the otherwise fcc Pd structure, increase the magnetic susceptibility. The (111) surface also increases the magnetic susceptibility and it becomes ferromagnetic in combination with an individual stacking fault or twin boundary close to it. On the contrary, we find that the (100) surface decreases the tendency to ferromagnetism. The results are consistent with the magnetic moment recently observed in small Pd nanoparticles, with a large surface area and a high concentration of two-dimensional stacking defects.

cond-mat.mtrl-sci↗

arXiv2005

Filtering a distribution simultaneously in real and Fourier space

We present a method to filter a distribution so that it is confined within a sphere of given radius r_c and, simultaneously, whose Fourier transform is optimally confined within a sphere of radius k_c. Our procedure may have several applications in the field of electronic structure methods, like the generation of optimized pseudopotentials and localized pseudocore charge distributions. As an example, we describe a particular application within the SIESTA method for density functional calculations, in removing the spurious rippling of the energy surface generated by the integrations in a real space grid.

physics.comp-ph↗

arXiv2004

Exchange and correlation as a functional of the local density of states

A functional $E_{xc}[ρ(\r,ε)]$ is presented, in which the exchange and correlation energy of an electron gas depends on the local density of occupied states. A simple local parametrization scheme is proposed, entirely from first principles, based on the decomposition of the exchange-correlation hole in scattering states of different relative energies. In its practical Kohn-Sham-like form, the single-electron orbitals become the independent variables, and an explicit formula for the functional derivative is obtained.

cond-mat.mtrl-sci↗

arXiv2003

Density functional simulation of small Fe nanoparticles

We calculate from first principles the electronic structure, relaxation and magnetic moments in small Fe particles, applying the numerical local orbitals method in combination with norm-conserving pseudopotentials. The accuracy of the method in describing elastic properties and magnetic phase diagrams is tested by comparing benchmark results for different phases of crystalline iron to those obtained by an all-electron method. Our calculations for the bipyramidal Fe_5 cluster qualitatively and quantitatively confirm previous plane-wave results that predicted a non-collinear magnetic structure. For larger bcc-related (Fe_35) and fcc-related (Fe_38, Fe_43, Fe_62) particles, a larger inward relaxation of outer shells has been found in all cases, accompanied by an increase of local magnetic moments on the surface to beyond 3 mu_B.

cond-mat.mtrl-sci↗

arXiv2001

Variational finite-difference representation of the kinetic energy operator

A potential disadvantage of real-space-grid electronic structure methods is the lack of a variational principle and the concomitant increase of total energy with grid refinement. We show that the origin of this feature is the systematic underestimation of the kinetic energy by the finite difference representation of the Laplacian operator. We present an alternative representation that provides a rigorous upper bound estimate of the true kinetic energy and we illustrate its properties with a harmonic oscillator potential. For a more realistic application, we study the convergence of the total energy of bulk silicon using a real-space-grid density-functional code and employing both the conventional and the alternative representations of the kinetic energy operator.

cond-mat.mtrl-sci↗

arXiv2001

Efficient index handling of multidimensional periodic boundary conditions

An efficient method is described to handle mesh indexes in multidimensional problems like numerical integration of partial differential equations, lattice model simulations, and determination of atomic neighbor lists. By creating an extended mesh, beyond the periodic unit cell, the stride in memory between equivalent pairs of mesh points is independent of their position within the cell. This allows to contract the mesh indexes of all dimensions into a single index, avoiding modulo and other implicit index operations.

cond-mat.mtrl-sci↗

arXiv2001

Evaluation of Exchange-Correlation Energy, Potential, and Stress

We describe a method for calculating the exchange and correlation (XC) contributions to the total energy, effective potential, and stress tensor in the generalized gradient approximation. We avoid using the analytical expressions for the functional derivatives of E_xc*rho, which depend on discontinuous second-order derivatives of the electron density rho. Instead, we first approximate E_xc by its integral in a real space grid, and then we evaluate its partial derivatives with respect to the density at the grid points. This ensures the exact consistency between the calculated total energy, potential, and stress, and it avoids the need of second-order derivatives. We show a few applications of the method, which requires only the value of the (spin) electron density in a grid (possibly nonuniform) and returns a conventional (local) XC potential.

cond-mat.mtrl-sci↗

arXiv2000

First principles study of the adsorption of C60 on Si(111)

The adsorption of C60 on Si(111) has been studied by means of first-principles density functional calculations. A 2x2 adatom surface reconstruction was used to simulate the terraces of the 7x7 reconstruction. The structure of several possible adsorption configurations was optimized using the ab initio atomic forces, finding good candidates for two different adsorption states observed experimentally. While the C60 molecule remains closely spherical, the silicon substrate appears quite soft, especially the adatoms, which move substantially to form extra C-Si bonds, at the expense of breaking Si-Si bonds. The structural relaxation has a much larger effect on the adsorption energies, which strongly depend on the adsorption configuration, than on the charge transfer.

cond-mat.mtrl-sci↗

arXiv2000

Zigzag equilibrium structure in monatomic wires

We have applied first-principles density-functional calculations to the study of the energetics, and the elastic and electronic properties of monatomic wires of Au, Cu, K, and Ca in linear and a planar-zigzag geometries. For Cu and Au wires, the zigzag distortion is favorable even when the linear wire is stretched, but this is not observed for K and Ca wires. In all the cases, the equilibrium structure is an equilateral zigzag (bond angle of 60$^{\rm o}$). Only in the case of Au, the zigzag geometry can also be stabilized for an intermediate bond angle of 131$^{\rm o}$. The relationship between the bond and wire lengths is qualitatively different for the metallic (Au, Cu and, K) and semiconducting (Ca) wires.

cond-mat.mtrl-sci↗

arXiv1999

Density functional calculations of planar DNA base-pairs

We present a systematic Density Functional Theory (DFT) study of geometries and energies of the nucleic acid DNA bases (guanine, adenine, cytosine and thymine) and 30 different DNA base-pairs. We use a recently developed linear-scaling DFT scheme, which is specially suited for systems with large numbers of atoms. As a first step towards the study of large DNA systems, in this work: (i) We establish the reliability of the approximations of our method (including pseudopotentials and basis sets) for the description of the hydrogen-bonded base pairs, by comparing our results with those of former calculations. We show that the interaction energies at Hartree-Fock geometries are in very good agreement with those of second order Møller-Plesset (MP2) perturbation theory (the most accurate technique that can be applied at present for system of the sizes of the base-pairs). (ii) We perform DFT structural optimizations for the 30 different DNA base-pairs, only three of which had been previously studied with DFT. Our results provide information on the effect of correlation on the structure of the other 27 base pairs, for which only Hartree-Fock geometries were formerly available.

physics.chem-ph↗

arXiv1999

Energetics of the oxidation and opening of a carbon nanotube

We apply first principles calculations to study the opening of single-wall carbon nanotubes (SWNT's) by oxidation. We show that an oxygen rim can stabilize the edge of the open tube. The sublimation of CO$_2$ molecules from the rim with the subsequent closing of the tube changes from endothermic to exothermic as the tube radius increases, within the range of experimental feasible radii. We also obtain the energies for opening the tube at the cap and at the wall, the latter being significantly less favorable.

cond-mat.mtrl-sci↗

arXiv1999

Stiff monatomic gold wires with a spinning zigzag geometry

Using first principles density functional calculations, gold monatomic wires are found to exhibit a zigzag shape which remains under tension, becoming linear just before breaking. At room temperature they are found to spin, what explains the extremely long apparent interatomic distances shown by electron microscopy.The zigzag structure is stable if the tension is relieved, the wire holding its chainlike shape even as a free-standing cluster. This unexpected metallic-wire stiffness stems from the transverse quantization in the wire, as shown in a simple free electron model.

cond-mat.mtrl-sci↗

arXiv1999

Linear-scaling ab-initio calculations for large and complex systems

A brief review of the SIESTA project is presented in the context of linear-scaling density-functional methods for electronic-structure calculations and molecular-dynamics simulations of systems with a large number of atoms. Applications of the method to different systems are reviewed, including carbon nanotubes, gold nanostructures, adsorbates on silicon surfaces, and nucleic acids. Also, progress in atomic-orbital bases adapted to linear-scaling methodology is presented.

cond-mat.mtrl-sci↗

arXiv1996

Selfconsistent order-N density-functional calculations for very large systems

We present a method to perform fully selfconsistent density-functional calculations, which scales linearly with the system size and which is well suited for very large systems. It uses strictly localized pseudoatomic orbitals as basis functions. The sparse Hamiltonian and overlap matrices are calculated with an $O(N)$ effort. The long range selfconsistent potential and its matrix elements are computed in a real-space grid. The other matrix elements are directly calculated and tabulated as a function of the interatomic distances. The computation of the total energy and atomic forces is also done in $O(N)$ operations using truncated, Wannier-like localized functions to describe the occupied states, and a band-energy functional which is iteratively minimized with no orthogonality constraints. We illustrate the method with several examples, including carbon and silicon supercells with up to 1000 Si atoms and supercells of $β$-C$_3$N$_4$. We apply the method to solve the existing controversy about the faceting of large icosahedral fullerenes by performing dynamical simulations on C$_{60}$, C$_{240}$, and C$_{540}$.

cond-mat↗

arXiv1995

Analysis of atomic-orbital basis sets from the projection of plane-wave results

The projection of the eigenfunctions obtained in standard plane-wave first-principle calculations is used for analyzing atomic-orbital basis sets. The "spilling" defining the error in such a projection allows the evaluation of the quality of an atomic-orbital basis set for a given system and its systematic variational optimization. The same projection allows obtaining the band structure and the Hamiltonian matrix elements in the previously optimized atomic basis. The spilling is shown to correlate with the mean square error in the energy bands, indicating that the basis optimization via spilling minimization is qualitatively equivalent to the energy-minimization scheme, but involves a much smaller computational effort. The spilling minimization also allows the optimization of bases for uses different than total-energy calculations, like the description of the band gap in semiconductors. The method is applied to the characterization of finite-range pseudo-atomic orbitals {[}O. F. Sankey and D. J. Niklewski, Phys. Rev. B {\bf 40}, 3979 (1989){]} in comparison to infinite-range pseudo-atomic and Slater-type orbitals. The bases are evaluated and optimized for several zincblende semiconductors and for aluminum. The quality of the finite-range orbitals is found to be perfectly comparable to the others with the advantage of a limited range of interacions. A simple scheme is proposed to expand the basis without increasing the range. It is found that a double-$z$ basis substantially improves the basis performance on diamond, whereas $d$-polarization is required for Si and Al for similar results. Finally, the projection allows the chemical analysis of the plane-wave results via population analysis on the previously optimized atomic basis.

cond-mat↗

arXiv1995

Projection of plane-wave calculations into atomic orbitals

The projection of the eigenfunctions obtained in standard plane-wave first-principle electronic-structure calculations into atomic-orbital basis sets is proposed as a formal and practical link between the methods based on plane waves and the ones based on atomic orbitals. Given a candidate atomic basis, ({\it i}) its quality is evaluated by its projection into the plane-wave eigenfunctions, ({\it ii}) it is optimized by maximizing that projection, ({\it iii}) the associated tight-binding Hamiltonian and energy bands are obtained, and ({\it iv}) population analysis is performed in a natural way. The proposed method replaces the traditional trial-and-error procedures of finding appropriate atomic bases and the fitting of bands to obtain tight-binding Hamiltonians. Test calculations of some zincblende semiconductors are presented.

cond-mat↗