The autoionizing states of doubly excited Rb atom were observed and their energy spectra measured in [1, 2]. In [1] the experimental spectra were registered at 75° polar angle with respect to the incident beam of electrons of 27 and 400 eV kinetic energy. This spectra are not fully identified until now. Theoretical studies of these states were made in [2, 3, 4]. So far, the most reliable published identification was made in paper [2]. More accurate attempts to identify this spectra are made in paper [4], where the first attempts to reliably identify the autoionizing states of doubly excited Rb were made by taking into account the asymmetry of the angular distribution of Auger electrons ejected from these states with respect to the incident beam of electrons. In the case of the spectra measured at various polar angles with respect to the incident beam of electrons of various kinetic energies, we are able to make a more fail-safe identification. There are published spectra measured at 75° [1] and 54.7° 'magic' [2] polar angles with incident beam of electrons of 27 and 400 eV; 16.7, 18.8 and 49.4 eV energy, respectively, which is only a few. There are no known published papers, where an experimental energy spectrum of autoionizing states of Rb was measured at a fixed polar angle with respect to the incident beam of electrons of more than three different kinetic energies, or vice versa. However, having in mind that the latter experimental spectra of papers [1, 2] were measured at different polar angles, we could simulate a theoretical ejected Auger electron spectrum for fixed parameters (angle of detection and incident electron energy) and compare the relative intensities of respective spectral lines in both spectra. By taking into account the asymmetry of angular distribution, we could relate most of the experimentally measured spectral lines to the theoretically calculated ones.
For the simulation of an Auger electron intensity spectrum we need to calculate the autoionization state energy, excitation cross section, radiative and autoionization probability values. It is known that the probability of a radiative decay is much smaller than the probability of an Auger decay from the autoionizing states of doubly excited Rb atom. Therefore, ejected Auger electron spectrum intensities of respective spectral lines are proportional to the excitation cross sections. Keeping in mind all the ideas mentioned previously we can write the differential cross section of a two step ionization process, where non-polarized atoms in the ground state are excited by non-polarized incident electrons to the doubly excited states as follows [5]:
$$\frac{d\sigma(\alpha_0J_0\to\alpha_1J_1\to\alpha_2J_2)}{d\Omega} = \frac{\sigma(\alpha_0J_0\to\alpha_1J_1\to\alpha_2J_2)}{4\pi} \left[1 + \sum_{K>0, \text{even}} \beta_K P_K(\cos\theta)\right]. \tag{1}$$Here $\sigma(\alpha_0 J_0 \to \alpha_1 J_1 \to \alpha_2 J_2)$ is the total cross section of a two step process, where $\alpha_0 J_0$ defines ground, $\alpha_1 J_1$ – excited states of an atom and $\alpha_2 J_2$ – ground state of an ion; $P_K(\cos\theta)$ is a Legendre polynomial; the asymmetry parameter of angular distribution of Auger electrons ejected from the autoionizing states of an atom is $\beta_K = A_K \alpha_K = A_K \cdot Const$ [6], because the Auger decay from these states of Rb atom is only possible to the ground state of Rb ion (decays to the excited states of Rb ion are energetically forbiden). This means that the angular distribution of Auger electrons ejected from the autoionizing states of Rb atom depends only on the alignment parameter $A_K$.
We have calculated that the asymmetry parameters $\beta_K$ for the autoionizing states 4p5nln'l' LSJ weakly depend on both orbital L and spin S angular momenta, but strongly depend on the total angular momentum J. Furthermore, we have shown that the angular distribution of Auger electrons ejected from the electron impact-excited autoionizing states could depend on the energy of impacting electrons even for the states with total angular momentum J > 3/2, when the respective spectral lines are measured at the 'magic' polar angle with respect to the incident electron beam.