The present Grasp2018 is an updated Fortran 95 version of the recommended block versions of programs from Grasp2K Version 1_1 for large-scale calculations Jönsson et al. (2013). MPI programs are included so that all major tasks can be executed using parallel computers. Tools have been added that simplify the generation of configuration state function expansions for the multireference single- and double computational model. Names of programs have been changed to accurately reflect the task performed by the code. Modifications to the relativistic self-consistent field program have been made that, in some instances, greatly reduce the number of iterations needed for determining the requested eigenvalues and the memory required. Changes have been made to the relativistic configuration interaction program to substantially cut down on the time for constructing the Hamiltonian matrix for configuration state function expansions based on large orbital sets. In the case of a finite nucleus the grid points have been changed so that the first non-zero point is Z-dependent as for the point nucleus. A number of tools have been developed to generate LaTeX tables of eigenvalue composition, energies, transition data and lifetimes. Tools for plotting and analyzing computed properties along an iso-electronic sequence have also been added. A number of minor errors have been corrected. A detailed manual is included that describes different aspects of the package as well as the steps needed in order to produce reliable results. Program summaryProgram Title:Grasp2018Program Files doi:http://dx.doi.org/10.17632/x574wpp2vg.1Licensing provisions: MIT licenseProgramming language: Fortran 95.Nature of problem: Prediction of atomic properties – atomic energy levels, isotope shifts, oscillator strengths, radiative decay rates, hyperfine structure parameters, specific mass shift parameters, Zeeman effects – using a multiconfiguration Dirac–Hartree–Fock approach.Solution method: The computational method is the same as in the previous Grasp2K [1,2] version except that only the latest recommended versions of certain routines are included.Restrictions: All calculations are for bound state solutions. Instead of relying on packing algorithms for specifying arguments of arrays of integrals, orbitals are designated by a “short integer” requiring one byte of memory for a maximum of 127 orbitals. The tables of reduced coefficients of fractional parentage used in this version are limited to sub-shells with j≤9∕2 [3]; occupied sub-shells with j>9∕2 are, therefore, restricted to a maximum of two electrons. Some other parameters, such as the maximum number of orbitals are determined in a parameter_def_M.f90 file that can be modified prior to compile time.Unusual features: Parallel versions are available for several applications.References•[[1]] P. Jönsson, X. He, C. Froese Fischer, and I. P. Grant, Comput. Phys. Commun. 176, 597 (2007).•[[2]] P. Jönsson, G. Gaigalas, J. Bieroń, C. Froese Fischer, and I. P. Grant, Comput. Phys. Commun. 184, 2197 (2013).•[[3]] G. Gaigalas, S. Fritzsche, Z. Rudzikas, Atomic Data and Nuclear Data Tables 76, 235 (2000).
Read full abstract