Year

2003

Degree Name

Doctor of Philosophy

Department

Institute for Superconducting and Electronic Materials - Faculty of Engineering

Abstract

Ag/Bi-2223 tapes doped with small quantities of highly enriched uranium were prepared by the powder-in-tube process and irradiated with thermal neutrons. The resulting fission fragments create randomly splayed quasi-columnar defects that provide strong flux pinning. Significant improvements in the field and angular dependence of the critical current density and in the irreversibility field were observed at 77 K, increasing with the density of fission tracks. In tapes with equivalent fission track density, the observed increases are larger for the samples exposed to larger neutron fluences (where a lower uranium doping level requires higher neutron fluence to obtain the same fission track density). This is explained as a result of the formation of secondary uranium-containing phases physically separated from the Bi2223 phase above a solubility limit to the uranium doping. An unusual exponential dependence of Hirr on the density of tracks was also observed, unrelated to the secondary phase formation. The primary investigation concerns the activation energies for vortex motion in the thermally activated flux flow and flux creep regimes, probed via resistive transition and dynamic magnetic relaxation measurements, which had not been previously explored. Both methods demonstrate the dominance of the fission tracks in the pinning landscape, particularly in fields applied parallel to the c-axis. Linearity in an Arrhenius plot of the superconductor resistance against temperature over at least three orders of magnitude indicated thermally activated flux motion in tapes with and without fission damage. In fields applied parallel to the c-axis, the pinning energy was substantially increased by the fission tracks, U0 tripling in magnitude at μ0H ~ 0.3 T, and then reconverging with the pre-irradiation U0 for μ0H > 2 T. This behaviour of the activation energy is well described by an effective matching field Beff ~ 0.3 T, identified at the maximum in the relative change of U0 and also observed from changes in the behaviour of the resistively determined irreversibility field. This Beff is lower than the matching field Bφ ~ 2.1 T estimated from the fission track density. Dynamic magnetic relaxation measurements were analysed using a modified form of Maley's method [M. P. Maley et al., Phys. Rev. B 42, p2639 (1990)]. This method was modified by using an empirically determined temperature scaling to resolve the current density dependence of the effective activation energy Ueff(J,H,T0) in the flux creep regime from different isothermal data points, over a wide range of temperatures and values of J. Correlated changes in the empirical temperature scaling form U(T), an additive constant C, a fitting parameter μ and the divergent behaviour of Ueff as J → 0 are all consistent with a dimensional crossover from 3D elastic to 2D plastic creep at fields μ0H ≈ 0.37 T in a nonirradiated tape. The dimensional crossover was shifted to higher fields μ0H ≈ 0.65 T after the introduction of fission-fragment damage, interpreted as evidence that the randomly splayed quasi-columnar defects promote c-axis vortex correlation in the Ag/Bi2223 tapes. Analysis of the relaxation data without the additional temperature scaling modification, on the other hand, produced a conflicting result, implying the complete destruction of c-axis vortex correlation over the entire field range explored. This observation is inconsistent with measurements of the current – voltage characteristics. Together with the unreasonable values of C obtained by this analysis and in comparison to the consistency observed with the empirically determined temperature scaling modification, the results indicate that the inclusion of the empirical temperature scaling is essential to an analysis of relaxation data. In comparison to past studies, we conclude that the efficiency of the uranium-fission method in enhancing the flux pinning is strongly dependent on the material anisotropy, through the vortex line elasticity and dimensionality, which hence affect the vortex line accommodation to the splayed defects. This has significant consequences for the estimation of Bφ in materials with differing anisotropy. The present method of calculating Bφ directly from the fission track density multiplied by the average defect track length is an oversimplification, which does not take into account the finite elasticity of the flux lines as well as the anisotropy of the vortex system.

Share

COinS
 

Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.