| There is a surprising lack of many body effects observed in XAS (X-ray Absorption Spectroscopy) experiments. While collective excitations and other satellite effects account for between 20% and 40% of the spectral weight of the core hole and photoelectron excitation spectrum, the only commonly observed many body effect is a relatively structureless amplitude reduction to the fine structure, typically no more than a 10% effect. As a result, many particle effects are typically neglected in the XAS codes used to predict and interpret modern experiments. To compensate, the amplitude reduction factor is simply fitted to experimental data.; In this work, a quasi-boson model is developed to treat the case of XAS, when the system has both a photoelectron and a core hole. We find that there is a strong interference between the extrinsic and intrinsic losses. The interference reduces the excitation amplitudes at low energies where the core hole and photo electron induced excitations tend to cancel. At high energies, the interference vanishes, and the theory reduces to the sudden approximation. The x-ray absorption spectrum including many-body excitations is represented by a convolution of the one-electron absorption spectrum with an energy dependent spectral function. The latter has an asymmetric quasiparticle peak and broad satellite structure. The net result is a phasor sum, which yields the many body amplitude reduction and phase shift of the fine structure oscillations (EXAFS), and possibly additional satellite structure. Calculations for several cases of interest are found to be in reasonable agreement with experiment. Edge singularity effects and deviations from the final state rule arising from this theory are also discussed.; The ab initio XAS code FEFF has been extended for calculations of the many body amplitude reduction and phase shift in x-ray spectroscopies. A new broadened plasmon pole self energy is added. The dipole matrix elements are modified to include a projection operator to calculate deviations from the final state rule and edge singularities. |
