Search for the "direct" channel in multiphoton double ionization of magnesium with femtosecond pulses

The removal of several electrons by multiphoton absorption from the outer shell of an atom is quite easy to observe with the currently available short pulse lasers. For instance, the outer shell of xenon atoms can be completely stripped out at 1016 Wcm -2. After the absorption of the first few photons in the discrete part of the spectrum, the atom can be (say) doubly ionized, in principle, through two quantum paths: either all the intermediate states of the multiphoton transitions are single-electron states (in this case, the electron correlations reflect only weakly in the process, and this results in the formation of the singly-charged ion in its ground state which, in turn can be ionized through only single-electron states etc…), or the multiphoton transition goes mostly through doubly-excited states up to the double ionization threshold (and is therefore more sensitive to electron correlations1). In the literature the first case is usually referred to as “sequential” and the second one “direct”. In the perturbation picture these names are clearly improper since, strictly speaking, multiphoton absorption is always a sequential process, only one photon being absorbed at each interaction. The more appropriate names of “correlated” and “uncorrelated” have been proposed.