Fig. 2: Fe Kβ main line x-ray emission.
a Fe Kβ main line difference spectra averaged in different time bins. Shaded areas reflect the standard deviation within a time bin when all difference spectra are rescaled to the bin-averaged summed difference signal magnitude. The black line (LMCT model) represents the ground state difference of aqueous 1[FeII(CN)6]4- and 2[FeIII(CN)6]3−. b Ligand-to-metal charge transfer excited state population (\({N}_{{LMCT}}\)) extracted from the fit of the main line region of the difference map shown in Fig. 1c. The uncertainty at each time delay was estimated using a cutoff in the increase of the sum of squared residuals when varying the population fraction with respect to the optimized value. c Population of the longer-lived species (\({N}_{S2}\)) extracted from the fit of the main line region of the difference map shown in Fig. 1c. The uncertainty at each time delay is estimated using a cutoff in the increase of the sum of squared residuals when varying the population fraction with respect to the optimized value. d The experimental 2[FeIII(CN)6]3−ground state spectrum (black) is shown together with calculated spectra for 2[FeIII(CN)6]3−, 4[FeIII(CN)6]3−, square pyramidal (SP) 2[FeIII(CN)5]2−, SP 4[FeIII(CN)5]2− and 2[FeIII(CN)5H2O]2−. All calculated spectra were aligned to their center of mass, then shifted by 1.1 eV to overlap the experimental and calculated 2[FeIII(CN)6]3− Kβ1,3 peaks. Peak positions of the calculated spectra are indicated by the vertical dashed lines. The peak positions of SP 4[FeIII(CN)5]2− and 2[FeIII(CN)5H2O]2− overlap.
