dftatom: A robust and general Schrödinger and Dirac solver for atomic structure calculations

Abstract A robust and general solver for the radial Schrodinger, Dirac, and Kohn–Sham equations is presented. The formulation admits general potentials and meshes: uniform, exponential, or other defined by nodal distribution and derivative functions. For a given mesh type, convergence can be controlled systematically by increasing the number of grid points. Radial integrations are carried out using a combination of asymptotic forms, Runge–Kutta, and implicit Adams methods. Eigenfunctions are determined by a combination of bisection and perturbation methods for robustness and speed. An outward Poisson integration is employed to increase accuracy in the core region, allowing absolute accuracies of 10−8 Hartree to be attained for total energies of heavy atoms such as uranium. Detailed convergence studies are presented and computational parameters are provided to achieve accuracies commonly required in practice. Comparisons to analytic and current-benchmark density-functional results for atomic number Z = 1 – 92 are presented, verifying and providing a refinement to current benchmarks. An efficient, modular Fortran 95 implementation, dftatom , is provided as open source, including examples, tests, and wrappers for interface to other languages; wherein particular emphasis is placed on the independence (no global variables), reusability, and generality of the individual routines. Program summary Program title: dftatom Catalogue identifier: AEPA_v1_0 Program summary URL: http://cpc.cs.qub.ac.uk/summaries/AEPA_v1_0.html Program obtainable from: CPC Program Library, Queen’s University, Belfast, N. Ireland Licensing provisions: MIT license No. of lines in distributed program, including test data, etc.: 14122 No. of bytes in distributed program, including test data, etc.: 157453 Distribution format: tar.gz Programming language: Fortran 95 with interfaces to Python and C. Computer: Any computer with a Fortran 95 compiler. Operating system: Any OS with a Fortran 95 compiler. RAM: 500 MB Classification: 2.1. External routines: Numpy ( http://www.numpy.org/ ) and Cython ( http://cython.org/ ) Nature of problem: Solution of the Schrodinger, Dirac, and Kohn–Sham equations of Density Functional Theory for isolated atoms. Solution method: Radial integrations are carried out using a combination of asymptotic forms, Runge–Kutta, and implicit Adams methods. Eigenfunctions are determined by a combination of bisection and perturbation methods. An outward Poisson integration is employed to increase accuracy in the core region. Self-consistent field equations are solved by adaptive linear mixing. Restrictions: Spherical symmetry Unusual features: Radial integrators work for general potentials and meshes. No restriction to Coulombic or self-consistent potentials; no restriction to uniform or exponential meshes. Outward Poisson integration. Fallback to bisection for robustness. Running time: For uranium, non-relativistic density functional calculation execution time is around 0.6 s for 10−6 a.u. accuracy in total energy on an Intel Core i7 1.46 GHz processor.

[1]  S. Fritzsche,et al.  On the role of the atom–cavity detuning in bimodal cavity experiments , 2007, 0711.2049.

[2]  Hamann Generalized norm-conserving pseudopotentials. , 1989, Physical review. B, Condensed matter.

[3]  W. J. Yost Self-consistent fields and diamagnetic susceptibility for magnesium III , 1940 .

[4]  G. Kresse,et al.  Relaxed core projector-augmented-wave method. , 2006, The Journal of chemical physics.

[5]  E. Cicely Ridley,et al.  The self-consistent field for Mo+ , 1955, Mathematical Proceedings of the Cambridge Philosophical Society.

[6]  I. Grant B-spline methods for radial Dirac equations , 2009 .

[7]  Barry N. Taylor,et al.  THE 1986 ADJUSTMENT OF THE FUNDAMENTAL PHYSICAL CONSTANTS: A REPORT OF THE CODATA TASK GROUP ON FUNDAMENTAL CONSTANTS , 1987 .

[8]  Frank. Tobin,et al.  The eigenvalue problem for a double minimum potential , 1975 .

[9]  Blöchl,et al.  Projector augmented-wave method. , 1994, Physical review. B, Condensed matter.

[10]  Sonnad,et al.  Purgatorio—a new implementation of the Inferno algorithm , 2005 .

[11]  J. Hinze,et al.  Nonrelativistic numerical MCSCF for atoms , 1986 .

[12]  D. Mayers,et al.  RELATIVISTIC ATOMIC WAVE FUNCTIONS. , 1971 .

[13]  W. Kołos,et al.  Theoretical Investigation of the Lowest Double‐Minimum State E, F 1Σg+ of the Hydrogen Molecule , 1969 .

[14]  C. S. Lin Theoretical analysis of the vibrational structure of the electronic transitions involving a state with double minimum: E, F 1 Σg+ of H2 , 1974 .

[15]  Mark D. Stiles,et al.  LOCAL-DENSITY-FUNCTIONAL CALCULATIONS OF THE ENERGY OF ATOMS , 1997 .

[16]  Weile Jia,et al.  The analysis of a plane wave pseudopotential density functional theory code on a GPU machine , 2013, Comput. Phys. Commun..

[17]  Emilio Artacho,et al.  The SIESTA method; developments and applicability , 2008, Journal of physics. Condensed matter : an Institute of Physics journal.

[18]  G. G. Stokes "J." , 1890, The New Yale Book of Quotations.

[19]  D. Andrae,et al.  Examination of several density functionals in numerical Kohn–Sham calculations for atoms , 2001 .

[20]  Charlotte Froese Fischer,et al.  A B-spline Galerkin method for the Dirac equation , 2008, Comput. Phys. Commun..

[21]  W. Johnson Lectures on Atomic Physics , 2014 .

[22]  J. E. Glynn,et al.  Numerical Recipes: The Art of Scientific Computing , 1989 .

[23]  P. Strange Relativistic Quantum Mechanics: With Applications in Condensed Matter and Atomic Physics , 1998 .

[24]  D. R. Hartree,et al.  The calculation of atomic structures , 1959 .

[25]  Dirk Andrae,et al.  Numerical electronic structure calculations for atoms. I. Generalized variable transformation and nonrelativistic calculations , 1997 .

[26]  I. P. Grant,et al.  The grasp2K relativistic atomic structure package , 2007, Comput. Phys. Commun..

[27]  Charlotte Froese Fischer,et al.  A general multi-configuration Hartree-Fock program , 1984 .

[28]  A. S. Douglas On the Sturm-Liouville equation with two-point boundary conditions , 1956, Mathematical Proceedings of the Cambridge Philosophical Society.

[29]  David J. Singh Planewaves, Pseudopotentials, and the LAPW Method , 1993 .

[30]  Peter Weinberger,et al.  Electron Scattering in Solid Matter , 2005 .

[31]  Jiří Vackář,et al.  Adaptability and accuracy of all-electron pseudopotentials , 2003 .

[32]  R. Martin,et al.  Electronic Structure: Basic Theory and Practical Methods , 2004 .

[33]  H. Skriver The LMTO Method , 1984 .

[34]  I. Grant The Dirac operator on a finite domain and the R-matrix method , 2008 .

[35]  D. Mayers Relativistic self-consistent field calculation for mercury , 1957, Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences.

[36]  J. Connerade,et al.  Controlling orbital collapse from inside and outside a transition element , 1998 .

[37]  C. Torrance Hartree Fields of Carbon , 1934 .

[38]  W. R. Fimple,et al.  A finite-difference Newton-Raphson solution of the two-center electronic Schrödinger equation , 1974 .

[39]  S. H. Vosko,et al.  Accurate spin-dependent electron liquid correlation energies for local spin density calculations: a critical analysis , 1980 .

[40]  G. E. Matthews,et al.  A Projector Augmented Wave (PAW) code for electronic structure calculations, Part II: Pwpaw for periodic solids in a plane wave basis , 2001 .