Costs of high-field superconducting strands for particle accelerator magnets

Abstract

The costs of superconducting magnet strands are compared by calculating a ‘production scaling factor’ P that relates purchase data to the cost of raw materials. Using a consistent method, we normalize for different conductor geometries and strand diameters to arrive at cost indices in $ kg−1, $ m−1, and $ kA−1 m−1. Analyses of Nb47Ti conductors taken from the past 25 years of high-field magnet projects reveal that the price of raw materials and, to a lesser extent, finished strands, have tracked the price of niobium pentoxide. Performance gains during the 1980s produced $ kA−1 m−1 indices that fell with time ahead of strand cost in $ m−1, a situation that may reflect the present status of Nb3Sn magnet conductors. Analyses of present materials show that P decreases systematically with billet mass. While production strands in 200–500 kg billets have costs times the cost of raw materials, the 20–50 kg billet size for internal-tin Nb3Sn composites drives P up to 8–10. Thus, in contrast to LHC-type Nb47Ti strands that cost $150 kg−1, $0.60 m−1, and $1.00 kA−1 m−1 at 5 T, 4.2 K, Nb3Sn strands required for the next generation of accelerator magnets are $1000 kg−1, $4.00 m−1, and >$5.75 kA−1 m−1 at 12 T, 4.2 K (where Jc is comparable to that for Nb47Ti at 5 T, 4.2 K). This high cost might be reduced by a factor of if a large-scale internal-tin or powder-in-tube Nb3Sn process can be found. Replacing expensive components with functionally equivalent but cheaper materials can produce 20% changes in $ kg−1 and $ m−1, but this might come at a performance penalty and no net savings in $ kA−1 m−1. Removing stabilizer from the strand cross-section and replacing it elsewhere in the cable can reduce the cost for a given length of cable significantly, but only if the processing cost for the strand remains unchanged after reduction of the stabilizer area. Emerging powder-in-tube composites show promise: Nb3Sn strands could reach $6 kA−1 m−1 at 12 T, 4.2 K, while Bi-2212 strands could fall below $10 m−1. MgB2 superconducting strands could have very low raw materials cost at $0.20 m−1, which translates to $1.00 m−1 for finished strands and at 2 T, 4.2 K based on published data.

ARIES-CS Magnet Conductor and Structure Evaluation

“…The cost of the superconducting material has been assumed to be $5000kg~2003 dollars!, slightly higher than some HEP experience 34,35 but substantially higher than what might be expected in the future. 33 The cost of the stabilizer~copper!…”

“…The largest cost of the conductor is due to the superconducting strands and the insulating and cabling. The cost of the superconductor strands and cable has been investigated by Cooley et al 34 and others. It has been hoped that the cost of the superconducting strands can decrease to as low as $1.50kA{m~12 T, 4.2 K!…”

Active compensation of field errors within ±2ppm in superconducting magnets

“…In the measurement conditions considered, the field variation over a turn generates an error up to ±1000 ppm of b 1 and ± 0.3 ppm of b 3 . The effect on the higher order harmonics can be neglected due to the high bucking ratio, about 1000, of the measurements coils.…”

“…In particular, the continuous demand for increasing collision energy has led to a remarkable research effort in superconducting magnets. Nowadays, the predominant material used in superconducting magnet coils of high-performance accelerators is NbTi, owing to its straightforward coil winding and cost effectiveness [2]- [3]. The advent of a reliable Nb 3 Sn technology will increase the maximum value of the generated field to beyond 10 T for future needs [4]- [5].…”

Study of High Field Superconducting Solenoids for Muon Beam Cooling

“…It was assumed that the relative volumetric costs of bare wires are: 1 m -3 for NbTi, 10 m -3 for Nb 3 Sn and 20 m -3 for HTS that is consistent with average present costs of superconducting materials [13]- [14].…”