Multielectron Inner-Shell Photoexcitation in Absorption Spectra of Krypton: Theory and Experiment

The probability of Kr 1s ionization alone and accompanied by excitation of a 4p, 3d, 3p, 2s, or a 2s and a 4p electron into bound or continuum states has been calculated for a photon-energy range extending a few hundred eV above the respective thresholds. The 1s ionization cross section was calculated relativistically including complete relaxation. Two-electron photoexcitation and ionization was calculated with Hartree-Fock wave functions, except for transitions to (1sns) final states, for which Dirac-Fock wave functions were employed. The calculations show that sharp features from two-electron excitations are produced in an absorption spectrum only if at least one electron undergoes a transition to a bound state. For more tightly bound electrons, shakeoff tends to prevail increasingly over shakeup, whence double -excitation features from inner-shell electrons become elusive. An absorption-spectrometry measurement was carried out with synchrotron radiation in order to test the calculations. Near the K edge of Kr, the measured absorption substantially exceeds the theoretical curve. For double-electron excitations, a background-subtraction technique permitted analysis of the 1s3d and 1s3p cross sections as far as ~ 130 eV above threshold. The observed slow rise in double-electron absorption due to shakeoff processes in 1s3d and 1s3p transitions agrees rather well with theory. A model which parameterizes the transition from the adiabatic to the sudden approximation regimes compares favorably with the 1s3d and 1s3p cross sections. Threshold cross sections for 1s4p, 1s3d, and 1s3p transitions are poorly predicted by the single-configuration Hartree-Fock calculations. For 1s2s photoexcitation, only a very slight change in slope of the total absorption cross section is predicted at threshold, which would make it difficult to measure with present-day facilities. The data exhibit an apparent edge in the 1s2s excitation region which, however, agrees with theory neither in energy nor in magnitude and is most likely due to extraneous effects. With improvements in techniques and in synchrotron-radiation sources, it may become possible in future to measure even these extreme inner-shell multielectron photoexcitation processes.