Large dynamic spin polarization parameters for diagram L₃M₁M_4,5Auger transitions

Abstract

Large dynamic spin polarization parameters have been calculated for at least three out of four lines which form the diagram L3M1M4,5 Auger spectra of the alkaline earth metals strontium, barium and radium, the transition metals palladium, zinc, cadmium and mercury, the rare gases krypton, xenon and radon, the lanthanide ytterbium as well as the actinide nobelium. The large values are primarily caused by the large phase shifts between the εp and εf partial waves of the emitted Auger electron which reduce the partial wave expansion of the dynamic spin polarization parameter to a sum of two or three dominant terms, only. Due to this, the possibility of a cancellation among the terms considerably decreases and, therefore, the entirely large values of the dynamic spin polarization parameters become understandable.

Comment on ‘Calculations of electron angular distribution in resonant Auger decay for Na, Ba, Hg and Kr*’

“…For the remainder of the comment we will leave any data obtained in the frozen-core approximation out and focus on Elizarov and Tupitsyn's realization that there is a huge discrepancy between their [10,11] and our [5] results for the angular anisotropy parameter α 2 for the L 3 M 1 M 4,5 Auger transitions of Kr, Xe, Ba, and Hg, see columns a and c of table 1. Again, these two data sets have been obtained by accounting for orbital relaxation during the Auger decay.…”

“…To ensure that our finding was not pure chance we extended our investigation to cover all L 3 M 1 M 4,5 Auger spectra of closedshell or closed-subshell atoms and found the same distinctive feature [5]. The calculations have been performed by describing the Auger emission process in the context of scattering theory (relativistic distorted wave approximation) where the Auger transition amplitudes and scattering phases have been evaluated by applying a relaxed orbital method within a multiconfiguration Dirac-Fock approach.…”

“…Expressions for the angular distribution and the dynamic spin polarization of Auger electrons featuring the above parameters can be checked in the literature (e.g. [5,6]). …”

“…The parity of the intermediate state and the final ionic state is odd and even, respectively. As the Coulomb operator conserves parity, the parity of the Auger electron and thus l and l ′ must be odd (for further explanation see [5]). Moreover, the numerator of equations (1) and (2) is invariant to the interchange of the two pairs of quantum numbers l, j and l ′ , j ′ which can be used to simplify the analysis of α 2 and ξ 2 .…”

“…In tables 2 and 3 we show all those phase differences which are essential for the calculation of α 2 and ξ 2 . The numerical values have been taken from our original work [5] (1) and (2) provided the Auger transition is to the final state 3 D 3 it is not necessary to consider this partial wave because the accompanying reduced Coulomb matrix element turns out to be zero in consequence of further selection rules arising from the angular part of the direct and exchange terms [8,9]. Tables 2 and 3 show that the phase differences between the partial waves with different orbital angular momenta l and l ′ are large whereas the phase differences between the partial waves with the same l are, without exception, considerably smaller.…”