The positron-hydrogen atom scattering system is considered within the S-wave model. Convergence in the elastic scattering, excitation, ionization, and positronium formation channels is studied as a function of the number and type of states used to expand the total wave function. It is found that all unphysical resonances disappear only if near-complete pseudostate expansions are applied to both the atomic and positronium centers.
34.85.+xIn recent years there has been substantial progress in the field of electron-atom collisions. Close-coupling based methods, utilizing near-complete pseudostate expansions about the atomic center, have been shown to yield very accurate results for discrete excitation and ionization channels, see for example [1][2][3][4]. This progress is primarily based on the study of the electron-hydrogen atom scattering S-wave model (only states with zero orbital angular momentum are retained) performed by Bray and Stelbovics [5]. They showed, by simply taking a pseudostate expansion whose completeness improves with increasing number of states N , that cross sections for discrete and ionization channels converged at all energies. Pseudoresonances, typically associated with small N calculations, disappeared for sufficiently large N , and convergence was to the correct independently evaluated results. This was an extraordinarily powerful result that is at the base of the substantial success recently enjoyed by the various implementations of the close-coupling theories.The situation for positron-atom scattering is somewhat more complicated. In addition to the atomically centered states the positronium formation channel needs to be included in the calculations in order to allow for all possible scattering channels. In other words, one needs a combined basis consisting of two independent basis sets. However, in this case if each of the components of the basis is large enough, one may expect instabilities in the calculations. The reason for this is that at small distances between colliding fragments functions of different basis components may essentially repeat each other due to their nonorthogonality. Thus, the closecoupling problem with the combined-basis-expansions * electronic address: Alisher.Kadyrov@flinders.edu.au is ill-conditioned. Is this an insurmountable obstacle? This question has remained unanswered for a long time. Mitroy, Berge, and Stelbovics [6] and Mitroy and Ratnavelu [7] have performed convergence studies for the full positron-hydrogen problem at low energies. Below the first hydrogen excitation threshold they showed good agreement between large pseudostate close coupling calculations and the highly accurate variational calculations of Humberston [8]. However, at higher energies, particularly above the ionization threshold the situation is less clear.We adopt the often-used CC(N, N ′ ) notation for closecoupling calculations that utilise N atomic eigenstates and N ′ positronium eigenstates to expand the total scattering wave function. In addition, a bar, when applied to N or N ′ ...