HTS Wire for Ultra-High Field Superconductor Magnets A. P. Malozemoff Consultant to AMSC, Devens MA AMSC | Windtec Solutions | Gridtec Solutions • Highest all-LTS-superconductor magnet: Bruker 23.5 T (1 GHz) NMR magnet — Near asymptotic limit for low temperature superconductors (NbTi/Nb3Sn) • HTS materials like YBCO have enormous (~100 T) upper critical field in low-T limit — Opens up a potentially huge new range of fields for science and NMR, using all-superconductor magnets • HTS wires are now commercially available, enabling such magnet systems All-SC magnet systems from 23.5 up to 50 T? • Exceed present highest DC field magnet: the 45 T hybrid at the National High Magnetic Field Lab (NHMFL) • Reduce operational cost (typical Bitter magnets cost $2000/hr. to run, or, for a year of continuous running $17.5 M!) — All-superconducting magnets would eliminate this dominating cost, expanding use dramatically • Avoid noise from forced water cooling of Bitter magnets • Enable revolutionary new science — Far more sensitive physics experiments probing quantum oscillations, decoherence, quantum computation, etc. — Far higher resolution in NMR for bioscience An impossible dream or credible opportunity? • Hoop stress σh = J x B x r • Assume σh = 500 MPa • Extrapolation to 50 T gives 4 cm bore, requires ~500 A/mm² under 50 T axial field [Figure 3: A calculation example of the current densities and peak field in the case of σh = 500 MPa. Number of coils = 10, Coil outer radius = 130 mm, Coil length = 520 mm, B0 = 36.5 T, E = 4.5 MJ. Otsuka & Kiyoshi, High Field Magnet Design Under Constant Hoop Stress, IEEE TAS, 2008] • HTS wires tape-shaped, with anisotropic field-dependence • Possible design will have: • 50 T axial field parallel to tape at inner bore • 15 T field perpendicular to tape at HTS coil ends [Diagram showing 20 T LTS outsert surrounding HTS insert providing 20 T + 30 T = 50 T axial field (parallel to tape orientation) and 15 T perpendicular field at ends] • Bi-2223 multifilament tapes (AMSC) Bi-2223 tapes exhaustively studied for power applications 30-77K: Now mature – lower Je than 2212 and YBCO • Bi-2212 round wire ~ 1mm dia. (Oxford SC) • REBCO coated conductor tapes (SuperPower) 4-12 mm wide by ~ 0.1 mm thick structure: - 20 µm Cu - 2 µm Ag - ~ 1 µm HTS - ~ 30 nm LMO - ~ 30 nm Homo-epi MgO - ~ 10 nm IBAD MgO - 50 µm Hastelloy substrate - 20 µm Cu Larbalestier - SciMag NRC Panel: Washington DC, March 12, 2012 • Processed wide, then slit to desired width • Laminated with copper, stainless… for strength, electrical stabilization • Cross-section: Copper laminates surrounding Substrate – Ni-5W and HTS – 1 µm YBCO layer (Total thickness: 200 µm) • Comparison showing Cu vs HTS 1000 A equivalents Well-established process producing 100’s km wire yearly • Amperium™ Copper Laminated 12 (~300 A/cm-width): - 3-layer, 10 mm wide HTS, 12 mm Cu - Average Ic = 296 A, σ = 3.7 A (1.2%) over 500+ m • Beta Wire – Double Insert (~550 A/cm-width effective): - 4-layer, 10 mm wide HTS, 12 mm Cu - Average Ic = 541 A, σ = 3.2 A (0.6%) over 200+ m High uniformity essential for stable operation e. g. SEI 1G (BSCCO) wire reinforced with 0.02-0.1 mm stainless steel Tensile strength @ 77 K (MPa) vs Ic retention (%): - SUS 0.10 mm maintains Ic retention up to ~500 MPa - SUS 0.05 mm maintains Ic retention up to ~380 MPa - SUS 0.02 mm maintains Ic retention up to ~270 MPa K. Sato et al., JJAP 51, 2012 K. Sato, SEI: “500 MPa of tensile stress will be feasible” Current Density Across Entire Cross-Section (Je [A/mm²], 4.2 K vs Applied Field [T]): - Je for 15 T perpendicular field: 500 A/mm² (YBCO B ⊥ Tape Plane) - Je for 50 T parallel field: 1300 A/mm² (YBCO B || Tape Plane) - Comparison data includes NbTi, Nb3Sn (Bronze and RRP), Bi-2212 round wire, Bi-2223 tape Note: BSCCO 2223 tape, round BSCCO 2212 wire possible future contenders. Larbalestier, NHMFL / SuperPower • AMSC – 2G wire: Ni-W substrate, MOD REBCO, laminated stabilizer • Bruker – 2G wire: Stainless steel substrate, PLD REBCO, laminated or electrodeposited stabilizer • Fujikura – 2G wire: Hastelloy substrate, PLD REBCO, electrodeposited stabilizer • Sumitomo Electric (SEI) – 1G wire: BSCCO/Ag composite, laminated stabilizer • SuperPower-Furukawa – 2G wire: Hastelloy substrate, MOCVD REBCO, electrodeposited stabilizer A growing HTS wire industry, but no standardization yet • Wire for ultra-high-field magnet systems still requires development • Optimization still needs to be done to further enhance Je at 4.2 K and ultra-high-field beyond today’s marginal values • Even higher strength tapes could be developed, e. g. by laminating to a thin tape of CuNb with yield strength > 1 GPa • No wire producer yet combines all the wire characteristics needed for a 50 T magnet • Cabling may be needed for some designs; industrial Roebel cabling still rudimentary • Are HTS wire producers interested in such a relatively small market, in comparison to the huge wire-volume opportunities in power cables, wind generators and fault current limiters? • Would they be willing to pursue the necessary development work under government contract? HTS wire already meets intermediate field requirements • Agilent (former Varian) – NMR systems • AMSC – all HTS magnet systems (e. g. 7 T research magnet) – now focused on coils for rotating machinery • Bruker – NMR systems with world’s highest field all-LTS system; also has BEST as HTS wire source; planning 28 T HTS NMR • Cryomagnetics – cryofree SC magnet systems • HTS-110 – so far mostly lower field HTS coils • Kobe Steel and JASTEC – cryofree SC magnet systems, NMR • Oxford Instruments – cryofree SC magnet systems, has worked with NHMFL on ultra-high-field HTS coil tests using Bi-2212 round wire • SEI – all HTS magnet systems (7 T research magnet) • Toshiba – cryofree LTS magnet systems, today focused on ITER, Maglev Are any of these ready to address hi-field HTS systems? S. Matsumoto et al., SUST, 2012 • 515 m of Fujikura GdBCO 2G wire in layer wound coil w/5 cm i. d. • Configuration: GdBCO insert coil installed in 17.2 T low-temperature superconducting (Nb3Sn and Nb-Ti) magnet. Highest all-superconductor field to date University of Houston & SuperPower • Je ~ 300 A/mm² • Stress levels 300 – 400 MPa • Coils tested: - SuperPower I: Bmax = 26.8 T, ΔB = 7.8 T - SuperPower II: Bmax = 27 T, ΔB = 7 T - NHMFL II: Bmax = 35.4 T, ΔB = 4.2 T K. Sato et al., JJAP 51, 2012 • Refrigerator-cooled Bi-2223 magnet system (20 K operation, 50 mm bore) • Double pancake coils wound with Bi-2223 wires SEI a leader in both HTS wire and its applications Kiyoshi et al., IEEE TAS 2010 • Uses SEI 1G “DI-BSCCO” wire • Coil configuration of the 1.03 GHz NMR magnet with Bi-2223 inner coil and Nb3Sn outer coils First operating NMR system w/HTS insert – target 2012 • High field 24/7 use: - Low operating cost - Breakthrough from current 18/20 T level • HTS Technology choices: - REBCO Coated conductor, single strand - Conservative design with margin • Focused project team: - Reliability is key - Supported by broader, aggressive R&D program • Close interaction with conductor vendor • The first ever all superconducting magnet with B > 24T • Key technological choices made; Now: full-featured test coils; 2013: User operations • Specifications: 32 T, 4.2 K, 32 mm bore; Standard "physics" homogeneity; Dilution refrigerator: <20 mK; HTS/LTS hybrid (LTS outsert from industry: Ø574mm, 745mm height; YBCO inner coil, Nb3Sn, NbTi) Larbalestier - SciMag NRC Panel: Washington DC, March 12, 2012 • Brookhaven National Lab HTS Magnet Program plans: • 24-30 T torus for ARPA-E SMES project (GRIDS SMES SYSTEM w/ SuperPower/University of Houston 2G HTS Wire, ABB Power Converter) • 40 T HTS solenoid test in 19 T resistive magnet background field: muon accelerator program (MAP) with Fermilab muon collider proposal • Grenoble High Magnetic Field Laboratory: • Interest in HTS for high fields, but no funded program yet • European funding for 5 kA HTS cables • LHC energy upgrade – CERN: • Goal ~20 T dipoles • HTS opens up the entire field range from 23.5 T to ~50 T for all-superconductor magnets • Multiple companies producing commercial HTS wire: Wire quality and quantity has matured to the point of enabling all-SC magnets >24 T, though more optimization required to achieve full range • Multiple companies have high field magnet systems expertise: Several already involved in initial ultra-high-field all-SC magnet systems • Government funding and co-operation with high field magnet labs like NHMFL or Tsukuba are likely needed International race is on to capitalize on this opportunity