Helical current flow along twist boundaries in YBCO thin films

Abstract

Current may flow in helical paths along the boundary between two tilted anisotropic conductors in circumstances when current flow in either conductor alone is restrained by the extreme anisotropy. We are able to explain number of features of the electrical properties of (103)/(013) oriented thin films deposited onto (110) substrates in terms of helical current paths involving the twist boundaries. These include the absence of weak link behaviour in the superconducting state for current flow along the substrate direction, the temperature dependence of the normal state resistivity for current flow in that direction, and the normal state electrical anisotropy. We predict that there are circumstances where currents will preferentially circulate around the boundaries of a tilt domain. The existence of these natural conduction loops may explain the period of recently observed quantum-interference behaviour in these films.

Anisotropic in-plane properties of (103)/(013) orientedYBa2Cu3OCampion1, King2, Benedict3 et al. 2000Phys. Rev. B7010View full textAdd to dashboardCite

“…They have an extremely low proportion of other phases such as the ͑110͒ phase and very low normal-state resistivities, [001] , comparable to the a-axis resistivity of a high-quality YBCO single crystal. The in-plane anisotropy values of the 0°films are in agreement with the meandering-current-path-helicalcurrent-flow model 38 which assumes that the 90°twist boundaries form a negligible impediment to current flow in the ͓001͔ direction, while enabling current flow in helical paths along the ͓11 0͔ direction. The model is supported by the anisotropies observed on very narrow tracks that inhibit circuitous current paths.…”

“…The degree of correlation in the ͓001͔ direction is important in the interpretation of the normal-state resistive anisotropy of these films. 38 In this structure domains of opposite tilt are connected along the ͓001͔ direction by 90°twist boundaries, and in the ͓11 0͔ direction by basal plane tilt boundaries or more commonly by symmetric or almost symmetric tilt boundaries. 25,37 These boundaries are illustrated in Fig.…”

“…Here 90°twist boundaries provide lobe matching and no severe disruption of the superconducting properties, while tilt boundaries provide mismatching and severe disruption. 38,39 Further disruption at tilt boundaries may be provided by crystallites of secondary phases, which are preferentially found at these sites. Tilting the cut of SrTiO 3 substrates by a few degrees away from the ͓110͔ towards the ͓11 0͔ direction causes the preferential growth of domains with one of the two alternative c-axis directions; the two tilts are no longer energetically equivalent.…”

“…Campion et al 38 have used this mechanism to propose that in the domained films which occur upon exact cut substrates ͓11 0͔ transport occurs by forward flow along the twist boundaries and sideways flow across ab planes, thus avoiding the tilt boundaries, which are expected to contribute considerable impediment to flow. Campion et al have been able to make a crude estimate of the normal-state anisotropy by estimating the effective length of the tortuous current paths involved on the plausible assumption that the domain tilts are uncorrelated in the ͓001͔ direction.…”

“…For films grown upon miscut substrates, such that one tilt occurs with a very low fraction, the current paths of Campion et al become very long and the shunting c-axis conduction dominates. 38 The observed in-plane anisotropy may then be deduced from the resistivity for a single crystal tilted by 45°with respect to the ͓110͔ crystallographic axis of the substrate: ϭ ab sin 2 (45°ϩ m )ϩ c cos 2 (45°ϩ m ) where m is the miscut angle, and ab is a suitable ''average'' of a and b . In the light of our 001 values we will use ab ϭ a , and c / ac ϭ120 ͑see Friedmann et al 40 predicted anisotropy is then 54 for the m ϭ3°and 49 for m ϭ5°.…”