Département de chimie moléculaire

First principle calcns. of extended x-ray absorption fine structure (EXAFS) data have seen widespread use in bioinorg. chem., perhaps most notably for modeling the Mn4Ca site in the oxygen evolving complex (OEC) of photosystem II (PSII). The logic implied by the calcns. rests on the assumption that it is possible to a priori predict an accurate EXAFS spectrum provided that the underlying geometric structure is correct. The present study studies the extent to which this is possible using state of the art EXAFS theory. The FEFF program was used to evaluate the ability of a multiple scattering-based approach to directly calc. the EXAFS spectrum of crystallog. defined model complexes. The results of these parameter free predictions are compared with the more traditional approach of fitting FEFF calcd. spectra to exptl. data. A series of seven crystallog. characterized Mn monomers and dimers was used as a test set. The largest deviations between the FEFF calcd. EXAFS spectra and the exptl. EXAFS spectra arise from the amplitudes. The amplitude errors result from a combination of errors in calcd. S02 and Debye-Waller values as well as uncertainties in background subtraction. Addnl. errors may be attributed to structural parameters, particularly in cases where reliable high-resoln. crystal structures are not available. Based on these studies, the strengths and weaknesses of using first-principle EXAFS calcns. as a predictive tool are discussed. A range of DFT optimized structures of the OEC may all be considered consistent with exptl. EXAFS data and caution must be exercised when using EXAFS data to obtain topol. arrangements of complex clusters. [on SciFinder(R)]