A theoretical description of dichroism in the photoelectron spectra of atoms polarized in different directions is presented. It is based on the general expression for the photoelectron angular distribution written in LSJ approximation. We show that, for photon energies far above threshold, the spectral distribution of dichroism in the photoelectron spectrum in most cases does not depend on the photoionization dynamics; for atoms with S symmetry, simple spectral patterns are obtained. Useful sum rules are given for the dichroism integrated over groups of photoelectron multiplet lines. Recent experimental data on magnetic dichroism and linear dichroism in the nonresonant inner-shell spectra of Cr (3p shell͒, Eu (4f shell͒, and Fe (3p shell͒ are analyzed. An outlook is given on the dichroism in the region of autoionization resonances. ͓S1050-2947͑99͒10808-4͔

No abstract

The study of the 5p photoemission from laser polarized ͑aligned or oriented͒ Eu atoms was used to deduce the ratio of the magnitude of the electric dipole amplitudes and the phase difference between the two outgoing ⑀s and ⑀d electron waves. The combination of magnetic dichroism experiments, the measurement of the difference spectra for two atomic polarizations, and phase tilt experiments, the observation of the modulation of the photoelectron signal on the angle of linear laser polarization, is a very promising method for performing a ''complete'' photoionization experiment. ͓S1050-2947͑98͒50111-4͔PACS number͑s͒: 32.80.Fb, 32.80.Hd RAPID COMMUNICATIONS R3374PRA 58 K. GODEHUSEN et al.

No abstract

The dichroism in the 4f photoelectron spectra of laser-aligned Eu atoms has been determined with linearly polarized vacuum ultraviolet radiation. The experimental results disagree with the prediction of the three-parameter model based on the absolute magnitude and relative phase of the 4f → εd, εg transition matrix elements. It is suggested that strong configuration interactions, confirmed in high-resolution photoelectron spectra, are the cause of the breakdown of the three-parameter model.