Skip to main content
. 2022 Jun 28;13(26):6164–6170. doi: 10.1021/acs.jpclett.2c01107

Figure 4.

Figure 4

(a) Molecular models for the structural components included in the Machine Learning (ML)-assisted EXAFS fitting model: 1 [CuI(NH3)2]+, 2 “planar”, and 3 “bent” motifs for μ-η22-peroxo diamino dicopper(II); when relevant, characteristic EXAFS-derived ranges for Cu–Cu interatomic distances are reported, in Å. (b–d) Comparison between magnitudes of experimental (colored circles) and best fit (thick lines) FT-EXAFS spectra at the end of the oxidation step for (b) 0.1_5, (c) 0.5_15, and (d) 0.6_29 (see SI, Figure S13 for the corresponding imaginary parts). The scaled components for Cu-species 1, 2/2′, 3 are also reported, vertically translated for the sake of clarity, together with percentages of each component over total Cu refined by ML-EXAFS fitting.