IOP PUBLISHING JOURNAL OF PHYSICS B: ATOMIC, MOLECULAR AND OPTICAL PHYSICS J. Phys. B: At. Mol. Opt. Phys. 42 (2009) 215204 (5pp) doi:10.1088/0953-4075/42/21/215204 Electron-impact double ionization of magnesium M S Pindzola 1 , J A Ludlow 1 , F Robicheaux 1 , J Colgan 2 and D C Griffin 3 1 Department of Physics, Auburn University, Auburn, AL, USA 2 Theoretical Division, Los Alamos National Laboratory, Los Alamos, NM, USA 3 Department of Physics, Rollins College, Winter Park, FL, USA Received 24 July 2009, in final form 9 September 2009 Published 27 October 2009 Online at stacks.iop.org/JPhysB/42/215204 Abstract Theory and experiment are compared for the electron-impact double ionization of Mg. Direct ionization cross sections, involving the simultaneous ionization of both 3s electrons, are calculated using a non-perturbative time-dependent close-coupling method. Indirect ionization cross sections, involving the ionization of either a 2p or 2s electron followed by autoionization, are calculated using a perturbative time-independent distorted-wave method. At low energies the direct ionization cross sections are found to be in good agreement with experiments, while at the higher energies the indirect ionization cross sections are also found to be in good agreement with experiments. (Some figures in this article are in colour only in the electronic version) 1. Introduction Electron-impact double ionization of atoms remains a challenging computational task for ab initio theory. Cross sections for light atoms are dominated by a direct process in which the incident electron stimulates the simultaneous emission of two electrons, resulting in a four-body Coulomb breakup problem. Work on the electron-impact double ionization of helium has found good agreement between non-perturbative time-dependent close-coupling calculations [1, 2] and absolute experimental measurements [3] for total cross sections. Recent reaction microscope experiments have provided energy and angle differential cross sections for the electron-impact double ionization of helium at incident energies just above threshold [4, 5]. Theoretical calculations based on a six interacting Coulomb waves method [4], a first Born implementation of the converged close-coupling method [4] and a time-dependent close-coupling method [6], have found good agreement between theory and experiment for pentuple differential cross section shapes, but not absolute magnitudes. On the other hand, electron-impact double-ionization cross sections for medium to heavy atoms are generally found to be dominated by an indirect process in which the incident electron ionizes an inner-shell electron, followed by autoionization of the excited atomic ion. Early theoretical and experimental studies of the electron-impact double ionization of inert gas atomic ions found that the indirect ionization– autoionization mechanism dominates total double-ionization cross sections [7]. Non-perturbative calculations for the electron-impact single ionization of loosely bound excited states of hydrogen [8, 9] and helium [10, 11] have shown that perturbative distorted-wave calculations are quite inaccurate. However, for the ionization of the tightly bound inner shell states in moderate to heavy atoms, the perturbative distorted- wave method should be reasonably accurate [12]. In this paper, we extend our recent non-perturbative calculations for the electron-impact single ionization of Mg [13] to examine the electron-impact double ionization of Mg. Although early experimental measurements [14, 15] of the double ionization cross section for Mg differed by a factor of 3, later experiments [16, 17] are in reasonable agreement with each other from threshold to 700 eV. From the double- ionization threshold of 22.7 eV to the 2p inner shell single- ionization threshold of 57 eV, we carried out calculations for the double ionization of both 3s electrons using the non- perturbative time-dependent close-coupling method. Above the 2p inner shell single-ionization threshold of 57 eV, we carried out calculations for the single ionization of the 2p and 2s electrons using the perturbative time-independent distorted- wave method. Comparisons are made with experiments [16, 17] from threshold to 500 eV. The remainder of the paper is organized as follows. In section 2, we review the time-independent distorted-wave 0953-4075/09/215204+05$30.00 1 © 2009 IOP Publishing Ltd Printed in the UK