Strand critical current density in Nb.SUB.3.Sn superconducting strands via a novel heat treatment

Abstract

A new heat treatment for Internal-Tin Nb.sub.3Sn strands is described. The heat treatment uses Nausite membranes to decrease the volume fraction of the phase and therefore minimize its liquefactionultimately resulting in better connected Nb.sub.3Sn. The heat treatment requires only one stage aside from the final Nb.sub.3Sn reaction stage. This heat treatment enables an increase in critical current density (at 16 T) of 28%.

Claims

1. A method for producing a high current density Nb.sub.3Sn superconducting wire comprising the steps of: a) fabricating a Cu clad internal tin Nb.sub.3Sn precursor wire containing Nb, Sn, Cu and a dopant of Ti and or Ta, drawn to an outer wire diameter between 0.2 mm and 2.0 mm, the precursor wire having a composition with an Nb fraction of 50-65% by area and an Sn fraction of 20-25% by area, with a Cu fraction representing the balance of the composition; and b) performing a heat treatment reaction wherein a Cu diffusion plateau between 350 C. and 380 C. is performed for 100 to 400 hours, thereby significantly reducing formation of (Nb.sub.0.75Cu.sub.0.25)Sn.sub.2 (Nausite), and a Nb.sub.3Sn reaction plateau is performed above 620 C. and below 750 C. for 24 to 400 hours.

2. A method in accordance with claim 1 wherein if the dopant element is Ta, it is part of the Nb fraction, with an alloy of Nb7.5 wt % Ta.

3. A method in accordance with claim 1 wherein, if the dopant element is Ti, it is part of Nb47 wt % Ti rods of 1-2% by area.

4. A method in accordance with claim 1, wherein the Nb.sub.3Sn reaction plateau is between 620 C. and 680 C. if the dopant is Ti.

5. A method in accordance with claim 1, wherein the Nb.sub.3Sn reaction plateau is between 650 C. and 700 C. if the dopant is Ta.

6. A method in accordance with claim 1, wherein the Nb.sub.3Sn reaction plateau time is between 50-400 hours.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) In the drawings appended hereto:

(2) FIG. 1 is an illustration of a distributed barrier Internal-Tin strand;

(3) FIG. 2 is micrograph of a partially reacted sub-element showing Nausite growth in the form of a membrane;

(4) FIG. 3 is a micrograph of fully reacted sub-elements showing disconnected Nb.sub.3Sn as a result of Nausite formation;

(5) FIG. 4 is micrograph of a partially reacted sub-element showing rapid Nausite growth after the liquefaction of the phase;

(6) FIG. 5 is a plot of Nausite membrane thicknesses as a function of time at various temperatures. Values marked with ** are predicted values that were later confirmed;

(7) FIG. 6 is a plot of Cu flux through the Nausite membrane;

(8) FIG. 7 is a plot of I.sub.c increase of a wire with a sub-element diameter of 35 m;

(9) FIG. 8 is a plot of copper content inside the sub-elements of a wire using the previous heat treatment and the proposed heat treatment.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

(10) For many years it has been noticed that, as shown in FIG. 3, some Nb.sub.3Sn 21 ends up disconnected from the main ring [12] and therefore cannot contribute to current transport. This disconnection is a result of the SnNbCu ternary phase, identified as (Nb.sub.0.75Cu.sub.0.25)Sn.sub.2 [13] and denominated Nausite, which forms at two instances during heat treatment:

(11) 1. The first formation of Nausite has the shape of a ring (or a membrane) that sits between the Cu/Nb filament pack and the CuSn core (as seen in FIG. 2). This Nausite membrane 20 is formed below 408 C. (the melting temperature of the CuSn phase or Cu.sub.6Sn.sub.5). When formed, the Nausite membrane enables Cu diffusion to the core while blocking Sn diffusion outwards. Such Cu diffusion is beneficial to consume the phase and produce more of the phase with a higher melting point. Unfortunately the membrane grows with time, producing more Nausite that results in disconnected Nb.sub.3Sn as previously stated. It does not appear of critical importance if this stage is preceded or not by a 210-215 C. Internal Tin Mixing dwell stage.

(12) 2. The second formation of Nausite 22 happens above 408 C. upon the liquefaction of the phase, and as shown in FIG. 4 it manifests in very large grains. The liquid corrodes the Nb filaments and produces large pieces of Nausite 23 rapidly which (as before) result in disconnected Nb.sub.3Sn.

(13) The novel heat treatment process described herein uses the Nausite membrane advantageously in order to diffuse the highest amount of Cu into the core (consuming as much as possible and therefore preventing liquefaction) while at the same time inhibiting the growth of the Nausite as a membrane.

(14) A difference between the novel heat treatment process described herein and those of the prior art described above is that the novel heat treatment process described herein doesn't focus on phase transformations or porosity. Rather, the present heat treatment process focuses on the formation of a Nausite membrane in order to facilitate the Cu diffusion to the core and prevent the liquefaction of as much as possible.

(15) The Nausite membrane can form in as little as 8 hours between 400 C. and 350 C. FIG. 5 shows the Nausite membrane thickness growth as a function of time. Since the Nausite phase produces disconnected Nb.sub.3Sn, lower temperatures are desired; however this membrane doesn't seem to grow continuously below 300 C. Once the membrane is formed, the Cu diffusion is activated, a diffusion which is weakly dependent on temperature between 400 C. and 370 C. as is shown in FIG. 6. It should be noted that the Cu diffusion slows down to impractical amounts below 350 C.

(16) The heat treatment process described hererin recommends the replacement of the previously thought as a Bronze Phase Mixing Dwell with a Cu Diffusion Dwell of 350-380 C. for >100 hours. For example, one dwell of 370 C. for 104 hours resulted in a Nausite layer thickness reduction of 46%. Due to the reduced formation of Nausite, as well as the increased Cu diffusion, this heat treatment has improved the critical current density (at 16 T) of the wires with sub-element size of 35 m by 27% as shown in FIG. 7. Furthermore, a 360 C. dwell for 150 hours results in a Nausite layer thickness reduction of 54% (critical current density not measured).

(17) Since at higher temperatures the Nausite membrane grows rapidly, the heat treatment dwell is recommended to be between 350 C. to 380 C., and for longer than 100 hours in order to diffuse as much Cu into the core to maximize conversion of phase bronze to phase bronze. Practical limitations of the dwell time are to a maximum of approximately 400 hours. In a preferred embodiment, the heat treatment process may feature a heat treatment of 370 C./104 hr+665 C./50 hrs.

Example 1

(18) For the purpose of proving the effectiveness of the heat treatment process, it was performed on a series of distributed barrier RRP strands [14]. These strands were manufactured from the same design to the same final diameter of 0.7 mm but nonetheless had a range in superconducting properties when heat treated with the standard heat treatment sequence: ramp from room temperature to 210 C. at a ramp rate of 25 C./hour, dwell of 210 C. for 48 hours, ramp to 400 C. at a ramp rate of 50 C./hour, dwell of 400 C. for 48 hours, and ramp to 650 C. at a ramp rate of 75 C./hour, dwell of 650 C. for 50-hours. FIG. 7 shows the Critical Current (I.sub.c) at 4.2K and in an applied magnetic field of 12 T for the standard heat treatment and the heat treatment with a modified Cu mixing dwell, ramp from room temperature to 370 C. at a ramp rate of 25 C./hour, dwell of 370 C. for 104 hours, ramp to 650 C. at a ramp rate of 75 C./hour, dwell of 650 C. for 50 hours. This particular wire saw an increase of 15% in critical current at 12 T and 27% increase in critical current at 15 T when the new heat treatment was applied. This heat treatment allowed producing a Nausite membrane that is 46% thinner than that produced in the standard heat treatment and increased the Cu content in the core by 9% as shown in FIG. 8.

(19) The exact ramp rate does not have a noticeable effect on the results of the heat treatment process. Ramp rates are generally chosen to be as fast as possible while still allowing for homogenous heating of the product; i.e. slower ramp rates are used for larger magnet coils. The practical range of ramp rates is 1 C./hour to 75 C./hour.

REFERENCES

(20) [1] A. R. Kaufmann and J. J. Pickett, Multifilament Nb.sub.3Sn Superconducting Wire, J. Appl. Phys., vol. 42, no. 1, pp. 58-58, January 1971. [2] C. van Beijnen and J. Elen, Multifilament Nb.sub.3Sn superconductors produced by the E.C.N. technique, IEEE Trans. Magn., vol. 15, no. 1, pp. 87-90, January 1979. [3] Y. Hashimoto, K. Yoshizaki, and M. Tanaka, Processing and properties of superconducting Nb.sub.3Sn filamentary wires, Proc. Fifth Int. Cryog. Eng. Conf. ICEC, no. Kyoto, 1974. [4] R. Schwall, G. Ozeryansky, D. Hazelton, S. Cogan, and R. Rose, Properties and performance of high current density Sn-core process MF Nb.sub.3Sn, IEEE Trans. Magn., vol. 19, no. 3, pp. 1135-1138, May 1983. [5] M. Field, J. Parrell, Y. Zhang, and S. Hong, Critical current density in Nb.sub.3Sn superconducting wire, U.S. Patent Publication 2006/0081307 A1, 20 Apr. 2006. [6] N. Higuchi, K. Tsuchiya, C. J. Klamut, and M. Suenaga, Superconducting Properties of Nb.sub.3Sn Multifilamentary Wires Fabricated by Internal Tin Process, in Advances in Cryogenic Engineering Materials, A. F. Clark and R. P. Reed, Eds. Springer US, 1984, pp. 739-746. [7] E. Gregory, G. M. Ozeryansky, and M. Suenaga, Some Effects of Porosity and Hip'ing on Critical Currents in Internal-Tin-Processed Multifilamentary Nb.sub.3Sn Wires, Brookhaven National Lab., Upton, N.Y. (United States), BNL-46846; CONF-91103061, January 1991. [8] S. Frtauer, D. Li, D. Cupid, and H. Flandorfer, The CuSn phase diagram, Part I: New experimental results, Intermetallics, vol. 34, pp. 142-147, March 2013. [9] S. Cogan, S. Kwon, J. Klein, and R. Rose, Fabrication of large diameter external-diffusion processed Nb.sub.3Sn composites, IEEE Trans. Magn., vol. 19, no. 3, pp. 1139-1142, May 1983. [10] A. McInturff and D. Larbalestier, Effect of metallurgical history on J.sub.c(5T) in surface diffused multifilamentary Nb.sub.3Sn, IEEE Trans. Magn., vol. 11, no. 2, pp. 259-262, March 1975. [11] E. Gregory, E. A. Gulko, and T. Pyon, Improvements in the properties of internal-tin Nb.sub.3Sn, IEEE Trans. Appl. Supercond., vol. 7, no. 2, pp. 1498-1503, June 1997. [12] J. A. Parrell, Y. Zhang, M. B. Field, P. Cisek, and S. Hong, High field Nb.sub.3Sn conductor development at Oxford Superconducting Technology, IEEE Trans. Appl. Supercond., vol. 13, no. 2, pp. 3470-3473, 2003. [13] S. Martin, A. Walnsch, G. Nolze, A. Leineweber, F. Laux, and C. Scheuerlein, The crystal structure of (Nb0.75Cu0.25)Sn2 in the CuNbSn system, Intermetallics, vol. 80, pp. 16-21, January 2017.

(21) [14] M. Field, J. Parrell, Y. Zhang, and S. Hong, Critical current density in Nb.sub.3Sn superconducting wire, U.S. Pat. No. 7,368,021 B2, 6 May 2008.