Most studies of Bi2Sr2CaCu2Ox (Bi2212) show that the critical current density J(c) is limited by the connectivity of the filaments, but what determines the connectivity is still elusive. Here we report on the role played by filament porosity in limiting J(c). By a microstructural investigation of wires quenched from the melt state, we find that porosity in the unreacted wire agglomerates into bubbles that segment the Bi2212 melt within the filaments into discrete sections. These bubbles do not disappear during subsequent processing because they are only partially filled by Bi2212 grains as the Bi2212 forms on cooling. Correlating the microstructure of quenched wires to their final, fully processed J(c) values shows an inverse relation between J(c) and bubble density. Bubbles are variable between conductors and perhaps from sample to sample, but they occur frequently and almost completely fill the filament diameter, so they exert a strongly variable but always negative effect on J(c). Bubbles reduce the continuous Bi2212 path within each filament and force supercurrent to flow through Bi2212 grains that span the bubbles or through a thin Bi2212 layer at the interface between the bubble and the Ag matrix. Eliminating bubbles appears to be a promising new path to raise the J(c) of Bi2212 round wires.

The decomposition of almost fully reacted (Bi, Pb)2Sr2Ca2Cu3Ox (BSCCO-2223) tapes caused by heating in 1 atm of pure O2 at 825 °C has been studied. It was found that partially decomposing 2223 tapes to a mixture of Bi2Sr2Ca1Cu2Oy, (Ca, Sr)2PbO4, and other secondary phases reduced the critical current density (77 K, 0 T) from ∼20 kA/cm2 to nearly zero. Reheating the tapes in 7.5% O2 restored the 2223 phase and, while there was some degradation of the 2223 grain alignment due to residual secondary phase growth, the critical current density was also restored to nearly its original value. We hypothesize that such a decomposition/reformation process can be useful in increasing the connectivity and relative density of polycrystalline 2223, by encouraging the formation of a liquid phase which heals residual cracks in the BSCCO core.

The evolution of Nb containing phases during the diffusion heat treatment of three different high critical current Nb3Sn strand types is compared, based on synchrotron X-ray diffraction results that have been obtained at the ID15 beam line of the European Synchrotron Radiation Facility (ESRF). In all strands studied, Nb3Sn formation is preceded by the formation of a Cu-Nb-Sn ternary phase, NbSn2 and Nb6Sn5. As compared to the PIT and Tube Type strand, the amount of these phases formed in the RRP strand is relatively small. In the RRP strand subelements with a fine filament structure Nb3Sn grows more quickly, thereby preventing to a large extent the formation of the other higher tin phases.

The evolution of Nb containing phases during the diffusion heat treatment of three different high critical current Nb3Sn strand types is compared, based on synchrotron X-ray diffraction results that have been obtained at the ID15 beam line of the European Synchrotron Radiation Facility (ESRF). In all strands studied, Nb3Sn formation is preceded by the formation of a Cu-Nb-Sn ternary phase, NbSn2 and Nb6Sn5. As compared to the PIT and Tube Type strand, the amount of these phases formed in the RRP strand is relatively small. In the RRP strand subelements with a fine filament structure Nb3Sn grows more quickly, thereby preventing to a large extent the formation of the other higher tin phases.

The push to drive superconductor strand technology to reach higher critical current density (J(c)) values and reduce production costs has led to innovative approaches in manufacturing technology. The Restacked Rod Process (RRP (R)) by Oxford Instruments is one such process which involves Nb bar extrusions in a Cu sheath. Commercially available Nb used in the initial RRP extrusion leads to nonuniform deformations of the Nb bar which in turn leads to a jagged Cu-Nb interface. This report presents a feasible methodology to remedy the problem of nonuniform deformation of Nb through severe plastic deformation (SPD) of precursor Nb to obtain smaller grains in starting Nb Cu-Nb monocore extrusion and drawing experiments were accomplished at Oxford Instruments using Nb bars of nominal dimensions 45mm diameter by and 78mm long and with grain sizes in the range of mu m to mm. Results of Cu-Nb interface roughness measurements show that a finer starting grain size gives a significantly lower roughness and better Nb core conformance to initial shape. Our experiments indicate that refinement of the initial Nb grain size to below similar to 50 mu m could enable fabrication of RRP conductor with improved wire yield.

The past several years have seen a significant improvement in the maximum critical current density (J(c)) in Nb3Sn strand. However for many applications, parameters besides high J(c) values are paramount. For fusion applications such as ITER, we have developed single-barrier internal tin strands having non-Cu J(c) values over 1000 A/mm(2) (12 T, 4.2 K) with hysteresis losses less than 1000 mJ/cm(3). Our most recent results are presented, with the goal of maintaining the high Jc but further reducing the losses. For high field magnet applications, higher Jc values are obtained using a distributed barrier approach. Results will be presented on a new high J(c) Nb3Sn strand that is made with (Nb,Ti)(3)Sn instead of (Nb,Ta)(3)Sn. This (Nb,Ti)(3)Sn strand has a J(c) value of 3000 A/mm(2) (12 T, 4.2 K), but has improved higher field performance compared with our standard (Nb,Ta)3Sn material, reaching 1700 A/mm(2) at 15 T, with further optimizations perhaps still possible. To reduce the effective filament diameter in these high J(c) strands, the number of subelement rods incorporated into the final restack billet needs to be increased from the 61 that was standard over the past several years. Results will be presented from manufacturing campaigns with 91 and 127 stack billet configurations.

The Restacked Rod Process (RRP) is the Nb{sub 3}Sn strand technology presently producing the largest critical current densities at 4.2 K and 12 T. However, when subject to plastic deformation, RRP subelements (SE) were found to merge into each other, creating larger filaments with a somewhat continuous barrier. In this case, the strand sees a larger effective filament size, d{sub eff}, and its instability can dramatically increase locally leading to cable quench. To reduce and possibly eliminate this effect, Oxford Instruments Superconducting Technology (OST) developed for FNAL a modified RRP strand design with larger Cu spacing between SE's arranged in a 60/61 array. Strand samples of this design with sizes from 0.7 to 1 mm were first evaluated for transport current properties. A comparison study was then performed between the regular 54/61 and the modified 60/61 design using 0.7 mm round and deformed strands. Finite element modeling of the deformed strands was also performed with ANSYS.

Performance improvements are needed for large scale applications of Nb3Sn, such as ITER or LHC upgrades. The highest critical current density (J(c)) values are achieved in distributed-barrier strand made by the Restacked Rod Process, which can reach 12 T, 4.2 K J(c) values of 3000 A/mm(2), with high residual resistivity ratio (RRR) values. For purposes of accelerator magnet stability, it is desirable to combine high J(c) with a small effective filament diameter (D-eff). Initial experiments show reducing D-eff from 80 mu m to 40 mu m leads to a 10% reduction in J(c). For fusion applications, a single-barrier design with well-spaced filaments is used to achieve the low hysteresis losses that are required. The status of our fusion strand development program is presented, including results for strand made using Nb-47 wt%Ti rods to supply Ti dopant. Such strands can reach 12 T, 4.2 K J(c) > 1000 A/mm(2), with losses < 1000 mJ/cm(3).

Final reactions of 48 or 50 hours were given in a range of temperature from 635 o C to 695 o C to high-J c Ta- and Ti-alloyed Restack-Rod-Process Nb 3 Sn strands to investigate the changes in critical current density and the superconducting properties. J c is the current density in the non-copper region of the wire. Measurements of J c were made at 4.2 K temperature and from 8 to 11.5 T field at BNL and from 12 to 16 T at OST, and these data were fitted to the Summers expression for J c to extrapolate the scaling field . B c2 * ldr All three(Nb, Ta) 3 Sn strand designs investigated displayed peak J c values of ~3000 A/mm 2 at 12 T and ~at 1500 A/mm 2 at 15 T for 665 and 680 o C reactions. B c2 * increased monotonically with increasing reaction temperature, a (Nb, Ti) 3 , By comparison, at each reaction temperature, a strand under development had a B c2 * higher than any of the Ta-alloyed strands. This suggests that Ti-alloyed strands could improve high-field performance if further development can bring their J c values closer to those of the Ta-alloyed strands. Other implications are also discussed.

Practical high field superconducting magnets are exclusively built with Nb3Sn multi-filamentary composites. Over the last few years there have been significant improvements in the current carrying capability of Nb3Sn strand, and these improvements offer the possibility to build more efficient and higher field strength magnets. The Nb3Sn composite requirements are somewhat different depending on the application, such as magnetically confined fusion, high energy physics accelerators, and solenoids for NMR or laboratory magnets, and thus require different designs. We will present the current status of Nb3Sn strand development at Oxford Instruments, Superconducting Technology (OST) for these applications, along with magnet results

AbstractChemInform is a weekly Abstracting Service, delivering concise information at a glance that was extracted from about 200 leading journals. To access a ChemInform Abstract, please click on HTML or PDF.

We continue development of Nb3Sn RRP® strand and BSCCO-2212 round wires for high field magnet applications. This year we studied RRP strand utilizing a new Ti doping method, accomplished by using a mixture of pure Nb and Nb-47wt%Ti filaments within the precursor (subelement) billet. The objective is to learn if high-field performance exceeding that of Nb-Ta alloy can be achieved using the Nb+Nb-Ti approach, by fine-tuning the Ti doping level.