Reduced Magnetization and Loss In Ag-Mg Sheathed Bi2212 wires: Systematics With Sample Twist Pitch and Length
MetadataShow full item record
Citation:Published version: IEEE Transactions on Applied Superconductivity (Volume: 25, Issue: 3, June 2015) Article Sequence Number: 8201604. https://doi.org/10.1109/TASC.2014.2372614
Suppression of magnetization and effective filament diameter (deff) with twisting was investigated for a series of recent Bi2212 strands manufactured by Oxford Superconducting Technologies. We measured magnetization as a function of field (out to 14 T), at 5.1 K, of twisted and non-twisted 37 x 18 double restack design strands. The samples were helical coils 5-6 mm in height and approximately 5 mm in diameter. The strand diameter was 0.8 mm. The magnetization of samples having twist pitches of 25.4, 12.7, and 6.35 mm were examined and compared to non-twisted samples of the same filament configuration. The critical state model was used to extract the 12 T deff from magnetization data for comparison. Twisting the samples reduced deff by a factor of 1.5 to 3. The deff was shown to increase both with L and Lp. Mathematical expressions, based upon the anisotropic continuum model, were fit to the data, and a parameter, γ2, which quantifies the electrical connectivity perpendicular to the filament axis, was extracted. The bundle-to-bundle connectivity along the radial axis was found to be approximately 0.2%. The deff was substantially reduced with Lp. In addition, the importance of understanding sample length dependence for quantitative measurements is discussed.
In this work we measured the suppression of magnetization and deff with twisting for an OST manufactured 0.8 mm 37 x 18 Bi2212 strand. Magnetization and deff values were suppressed by factors of 1.5-3, making deff and magnetization 1.5-3 times smaller for twisted samples as compared to nontwisted samples. This effect was further systemized and quantified by looking at the dependence of deff on Lp, and also the dependence of deff on L. A model was applied which described the linear dependence on both L and Lp, and extracted a value for the connectivity parameter 2; a value of only 0.2% was found between the subelements. We conclude that (1) loss, magnetization, and deff are suppressed by sample twisting, (2) it is possible to quantify this effect by a parameter 2, and (3) it is important to have long samples (L >> Lp, and also L > Lcrit) to obtain results most relevant to application.
This work was supported by the U.S. Department of Energy, Office of Science, Division of High Energy Physics, under Grant DE-SC0010312 and DE-SC0011721