X-ray absorption fine structure characterization of the local structure of Fe in Fe-ZSM-5

The local structure of Fe in Fe-ZSM-5 prepared by solid-state exchange was investigated with XAFS. Fe K-edge spectra were taken at liquid nitrogen temperature for samples with Fe/Al ratios of 0.33, 0.66, and 0.80. The radial structure function (RSF) of He- and CO-pretreated Fe-ZSM-5 shows two main peaks, one at 1.6 Angstrom and the other at 2.5 Angstrom. To interpret the origin of these peaks, RSFs were simulated for a number of mono- and di-iron structures obtained from quantum chemical calculations. By this means, the peak in the RSF at 1.6 Angstrom is clearly identified with backscattering from 0 atoms coordinated to an Fe atom. The peak at 2.5 Angstrom has been previously ascribed to Fe-Fe scattering and has been used to argue for the presence of di-iron-oxo species; however, the origin of this peak and its interpretation remains an open question. The imaginary part of the Fourier transformed data for the peak at 2.5 Angstrom has the same characteristics as that generated theoretically for Fe-Al backscattering and is distinctly different from that generated theoretically for Fe-Fe backscattering. This evidence strongly suggests that the iron in Fe-ZSM-5 is present as isolated cations associated with framework aluminum. Further evidence for such a structure is the absence of any change in the magnitude of the peak near 2.5 A with sample treatment. The RSFs and the information obtained from curve-fitting demonstrate that the structure of Fe in Fe-ZSM-5 does not change significantly with Fe/Al ratio. For both He- and CO-pretreated sample, the Fe-O coordination number is about 4 and correspondingly the Fe-Al coordination number is about 1, regardless of Fe/Al ratio. Therefore, the structure of Fe in Fe-ZSM-5 is best described as either Z(-)[Fe(O)(2)](+) or Z(-)[Fe(OH)(2)](+), where Z- represents the charge-exchange site in the zeolite. Upon O-2 pretreatment, a new feature appears at about 1.1 Angstrom in the RSF, which may be due to a migration of some of the Fe into the zeolite framework. This interpretation is qualitatively consistent with the observed RSF for Fe-silicalite.