Strong interference between decay channels and valence-electron rearrangements in core-hole spectroscopy

In the analysis of core-hole spectra it is usually assumed that the response of the valence electrons upon the creation of a core hole can be treated independent of the subsequent decay of the core hole. We show that for near-threshold decay processes this is no longer true and that the modification of the satellite spectra is considerable even for relative slow decay processes. Taking the 2p x-ray photoemission spectra of ${\mathrm{NiBr}}_2$ as an example, we show that the large difference between the 2$p_{3/2}$ and 2$p_{1/2}$ spectra of transition-metal compounds is due to the near-threshold Coster-Kronig decay of the 2$p_{1/2}$ hole.