N.S. Scott

Electron scattering by atoms and ions using the Breit-Pauli Hamiltonian: An R-matrix approach

The R-matrix method describing the scattering of low-energy electrons by complex atoms and ions is extended to include terms of the Breit-Pauli Hamiltonian. An application is made to the astrophysically important 1s22s2 1S0e-1s22s2p 3P1o transition in Fe XXIII, where in the most accurate calculations carried out all terms of the 1s22s2, 1s22s2p and 1s22p2 configurations are included in the expansion describing the collision. This gives up to 28 coupled channels for each total angular momentum and parity which are solved on a CRAY-1. The collision strengths are increased by more than a factor of two from their non-relativistic values at all energies considered.

The scattering of low-energy electrons by argon atoms

Phaseshifts, differential, total and momentum transfer cross sections are calculated using an R-matrix approach for the elastic scattering of electrons by argon atoms in the impact energy range 0-19 eV. The coupled-state calculation is based upon a single-configuration atomic ground-state wavefunction coupled to a 1P pseudostate. A critical assessment of earlier theoretical and experimental data is made and the conclusion is reached that the present results are the most satisfactory over the entire energy range considered.

Numerical 'health check' for scientific codes: the CADNA approach

Scientific computation has unavoidable approximations built into its very fabric. One important source of error that is difficult to detect and control is round-off error propagation which originates from the use of finite precision arithmetic. We propose that there is a need to perform regular numerical ‘health checks’ on scientific codes in order to detect the cancerous effect of round-off error propagation. This is particularly important in scientific codes that are built on legacy software. We advocate the use of the CADNA library as a suitable numerical screening tool. We present a case study to illustrate the practical use of CADNA in scientific codes that are of interest to the Computer Physics Communications readership. In doing so we hope to stimulate a greater awareness of round-off error propagation and present a practical means by which it can be analyzed and managed.

A new no-exchange R-matrix program

Abstract An R -matrix program is described implementing a new faster technique for calculating the elements of the Hamiltonian matrix when exchange and the quadratically integrable functions are omitted. It gives accurate results for high partial waves and can be used to supplement low partial wave results including exchange.

2DRMP: A suite of two-dimensional R-matrix propagation codes

The R-matrix method has proved to be a remarkably stable, robust and efficient technique for solving the close-coupling equations that arise in electron and photon collisions with atoms, ions and molecules. During the last thirty-four years a series of related R-matrix program packages have been published periodically in CPC. These packages are primarily concerned with low-energy scattering where the incident energy is insufficient to ionise the target. In this paper we describe 2DRMP, a suite of two-dimensional R-matrix propagation programs aimed at creating virtual experiments on high performance and grid architectures to enable the study of electron scattering from H-like atoms and ions at intermediate energies.

The 6s6p2 resonances in e-Hg scattering

The elastic scattering and excitation cross sections for the (6s6p)3P0,1,20 states in e-Hg scattering are calculated from 4-7 eV using the relativistic R-matrix method. Resonances corresponding to the 6s6p2 configuration are found and their effect on the cross sections is discussed. In most cases good agreement with experiment is obtained, and certain features have been clarified.

Graphical R-matrix atomic collision environment (GRACE): the problem specification stage

Abstract In this paper we introduce the concept of a graphical R-matrix atomic collision environment (GRACE). GRACE couples the graphical capability of powerful workstations with the processing power of supercomputers to provide an environment for the study of atomic collision properties and processes. At the core of GRACE is a new generation R-matrix program package, which is used to compute properties characterising electron atom and electron ion collisions. One of the motivations behind the project is to render this package simple to use by novice and experienced users alike, thereby significantly improving its usefulness to the physics community. GRACE is composed of a problem specification stage, a computation stage, and an interpretation stage. The focus of this paper is a description of the X Window graphical user interface which constitutes the problem specification stage of GRACE.

Computational aspects of the two-dimensional propagation of R-matrices on MPPs

Abstract The simulation of electron collisions, at intermediate energies, with atoms, ions and molecules in highly excited Rydberg states is a physically important and computationally challenging problem. A suitable computational technique, to meet this challenge for the general two-electron problem, involves an adaptation of the intermediate energy R -matrix approach, in which the ( r 1 , r 2 ) plane of the inner region is partitioned into a set of connected subregions and a global -matrix is propagated, in two dimensions, across this plane. In this paper we examine the computational aspects of this technique and, in particular, describe a parallel strategy for exploiting the architecture of distributed memory parallel computers.

An array processing language for transputer networks

Abstract Networks of transputers are often used for array processing applications. Programming such systems in OCCAM can be tedious and error-prone. This paper outlines a language designed to facilitate the solution of problems which involve some array processing. The language is called LATIN, and is currently being implemented on a transputer network. An example of its use is given.

Fast Computation of the Slater Integrals

Slater integrals are two-dimensional radial integrals whose integrand is constructed from normalized eigenfunctions of the Schrodinger equation. These integrals occur in many atomic structure and scattering computations. However, in two-dimensional R-matrix propagation they represent a significant computational bottleneck. The problem involves two steps, that is, the numerical solution of the Schrodinger equation and the computation of the integrals, respectively. By exploiting the characteristic features of the problem we seek to devise a two-stage computational strategy, where the second stage is influenced and informed by the first. In particular, we focus on the development of extended frequency-dependent quadrature rules (EFDQRs) that improve the accuracy and significantly reduce the computation time of the integrals. The performance of these ad hoc rules is examined and compared to the currently used Newton--Cotes method in the construction of Hamiltonian matrices involving up to 300 x 106 integrals. A gain of two orders of magnitude in the CPU time is obtained.