This paper describes a new technique to quantitatively extract the number of unoccupied d states in a material utilizing measurements of the L X-ray absorption edge spectra. A correlation between the area under each of the L2 and L3 X-ray absorption edges and d-band vacancies in platinum-containing materials which exhibit white lines is given for the first time. The technique is demonstrated with a platinum catalyst supported on silica as an example. The quantity determined is the fractional change of the d-band occupancy for the sample from that of bulk platinum.

No abstract

Four highly dispersed and fully reduced rhodium on alumina catalysts with different particle sizes in the range 6–12 Å were investigated with the EXAFS technique in order to derive information about the structure of the metal–support interface. This information can only be obtained when the signal-to-noise ratio of the experimental EXAFS data is high enough and accurate reference compounds and a modified way of data analysis are used. With the aid of phase and amplitude corrected Fourier transforms it was possible to detect a small additional signal which could be ascribed to a Rh–O bond. Since the catalysts were fully reduced and since the intensity of the small signal increased with decreasing particle size, the oxygen neighbor was assigned to be originated from the metal–support interface. From the intensity of the Rh–O bond it was estimated that, on the average, each interfacial rhodium atom is surrounded by 2–3 oxygen ions of the support. The detected Rh–O bond has a coordination distance of 2.7 Å which is about 0.6 Å larger than the first coordination distance in Rh2O3 (2.05 Å). The coordination distance of 2.7 Å can be explained by assuming an interaction between metallic rhodium (atomic radius 1.34 Å) and ionic oxygen belonging to the support (ionic radius 1.4 Å). This would possibly imply an ion-induced dipole bonding between the metal particle and the support.

In capacity measurements, a strong frequencydependent capacity has been observed at higher potentials which was discussed as a contribution of surface states.8In photocurrent spectra, a potential-dependent onset energy similar to the one described for thin films on zirconium is observed at low potentials. This is another indication that cathodic photocurrents are important at low potentials. The experimental results for passive iron, however, are not yet as detailed as for the other systems, allowing for no quantitative interpretation. ConclusionsDetailed results of photocurrent measurements on passive films revealed the presence of cathodic photocurrents at potentials above the flatband potential. The potentials, U0, at which the photocurrent changes sign are found to be wavelength dependent which indicates that U0 is not the flatband potential. It is also found that UQ is usually more positive than Uvalues obtained from capacity measurements. Passive films are generally highly disordered or amorphous. This means that a large number of localized states are present in what is the bandgap of the corresponding crystalline material. For a photoexcitation process that involves localized states, the removal of the photoexcited charge carrier from the localized state can be rate determining.1 2The model presented here takes into account reverse tunneling, i.e., tunneling processes against the field, which can result in cathodic photocurrents. The model calculations generate results that are very similar to experimental results and which are able to explain experimental results described above.

Synchrotron based mammography imaging experiments have been performed with monochromatic x-rays in which a laue crystal placed after the object being imaged has been used to split the beam transmitted through the object. The X27C R&D beamline at the National Synchrotron Light Source was used with the white beam monochromatized by a double crystal Si(1 11) monochromator tuned to 18keV. The imaging beam was a thin horizontal line approximately 0.5mm high by 100-wide. Images were acquired in line scan mode with the phantom and detector both scanned together. The detector for these experiments was an image plate. A thin Si(1 1 1) laue analyzer was used to diffi.act a portion of the beam transmitted through the phantom before the image plate detector. This "scatter fiee" diffracted beam was then recorded on the image plate during the phantom scan. Since the thin laue crystal also transmitted a hction of the incident beam, this beam was also simultaneously recorded on the image plate. The imaging results are interpreted in terms of an x-ray schliere or rehctive index inhomogeneities. The analyzer images taken at various points in the rocking curve will be presented.