Programs in Physics & Physical Chemistry
|[Licence| Download | New Version Template] acln_v1_0.gz(400 Kbytes)|
|Manuscript Title: RMATRX-ION: a program to calculate electron and positron impact ionization within the R-matrix method.|
|Authors: K. Bartschat|
|Program title: RMATRX-ION|
|Catalogue identifier: ACLN_v1_0|
Distribution format: gz
|Journal reference: Comput. Phys. Commun. 75(1993)219|
|Programming language: Fortran.|
|Computer: VAX 8650.|
|Operating system: VMS.|
|RAM: 735K words|
|Word size: 64|
|Keywords: Atomic physics, Electron, Scattering, Photoionization, Electron and positron Impact ionization, Close coupling, R-matrix, Resonances coulomb-born Approximation, Ls coupling.|
Nature of problem:
This program can be used to calculate the single differential (with regard to energy loss) and total (i.e., integrated over all energy losses) cross sections for impact ionization of atoms and ions by electrons and positrons.
A "fast" incident projectile is described by a distorted wave while the interaction of a "slow" ejected electron with the residual ion is accounted for by an R-matrix (close-coupling type) expansion.
Exchange effects between an ionizing and an ejected electron and the possibility of positronium formation between an ionizing positron and an ejected electron are neglected, i.e., the process is described in the "Distorted Wave Coulomb Born No-Exchange" approximation.
Like previous versions of the R-matrix codes, the program is designed in several stages. Due to the energy dependence of the operator describing the Coulomb interaction between the incident projectile and the target, the program should be run for the desired energies of the ejected electron at several "key energy" losses of the incident projectile. The value at the correct energy loss can then be obtained through interpol- ation. Furthermore, the built-in restart facilities of the R-matrix codes are used to ensure that the "inner region" of the (N+1)-electron problem (ejected electron and residual ion) is only treated once.
The running time depends very strongly on the actual problem. The test run up to the first run through MCONST took approximately 57 minutes of CPU time on a VAX 8650 (4 seconds in POT, 14 seconds in IONDWBA, 31 minutes in STG1, 5 minutes in STG2, 15 minutes in STG3 (DIAG) and 5 minutes in MCONST). The restart for this part (IONDWBA, STG1 and MCONST) at a different key energy loss took 35 minutes, since only part of STG1 had to be run. The asymptotic part for two energy losses and six final state symmetries took 3 minutes of CPU-time per key energy loss. In order to check the interpolation program INTPOL, another test run for 149 energy losses was performed which took 157 minutes of CPU- time in the asymptotic part for the first key energy loss; the restart of IONDWBA, STG1, MCONST and STG3 (ASYM) for a different key energy loss with exactly the same ejected electron energies needed 61 minutes of CPU-time key energy loss. Finally, the interpolation procedure in INTPOL using the results for four such key energy losses and 149 different energies took 11 seconds. It should be noted that the asymptotic part can be speeded up significantly by using more recently developed asymptotic packages of the Belfast/Daresbury groups. However, since these programs have not been published yet, the present code is provided with the published asymptotic package of Crees.
|Disclaimer | ScienceDirect | CPC Journal | CPC | QUB|