No abstract

No abstract

No abstract

Iron diffusion in copper has been measured using the Mossbauer effect. The temperature dependence of the Mossbauer line broadening is proportional to the temperature dependence of the diffusivity as determined by the tracer-sectioning technique. The magnitude of the broadening is about h/r (where r is the mean time an atom spends at a given site), instead of Ih/r as predicted by the jump model of diffusion. No evidence for sidebands is observed in the broadened Mossbauer spectra, which have approximately Lorentzian line shape. The temperature dependence of the recoil-free fraction appears to be anomalously large in the region where diffusion broadening of the Mossbauer line was observed.S HORTLY after the discovery of the Mossbauer effect, theoretical arguments 1 were given to show how the effect might be used as a microscopic probe for studying diffusion in solids. Although earlier experimenters 2,3 reported the observation of diffusion associated with the Mossbauer effect, these data were handicapped by the indirect way in which the linewidth was measured, they failed to show the expected temperature dependence 1 for the broadened line, and they were not convincingly correlated to macroscopic diffusion experiments. By contrast, other observers have sought direct observations of diffusion broadening but they did not observe the expected effect. 4 We wish to report the direct observation of diffusion broadening of the Mossbauer line of Fe 57 in copper. We have observed the temperature dependence of this broadening, and, within present experimental errors, the increase in linewidth does follow the exponential temperature dependence observed in the usual type of diffusion measurement, which is in accord with theory, although the magnitude of the broadening appears to be only about half of the amount predicted by the simple diffusive jump model originally considered in relation to the Mossbauer effect by Singwi and Sjolander. 1 In harmony with their analysis, we find no evidence of side peaks in the wings of the resonance as has been observed in time filtering experiments. 5,6 The data were taken with a crank velocity spectrometer system. 7 As a check on possible systematic errors, a source experiment using Co 57 in Cu versus a Na4Fe(CN)6*10H 2 O absorber was used for one set of measurements, while an Fe 57 -doped copper absorber scanned with a Co 57 -in-Pd source was used for the other *

No abstract