Nb₃Sn is the workhorse superconductor for high-field magnets — particle accelerators, fusion reactors, MRI machines. But it is brittle. You cannot weld it, bend it, or join pieces without destroying the superconducting phase at the junction. Every magnet must be wound from a continuous conductor, which constrains the manufacturing geometry.
This paper demonstrates continuous supercurrent across a diffusion-bonded seam in Nb₃Sn thin films. Two bronze pieces are joined by diffusion bonding at approximately 715°C, then Nb is deposited by vapor and reacted to form the superconducting phase across both pieces and the seam. At 9 K, supercurrent flows freely through the junction.
Note (2026-11-08): the abstract (Juliao et al.) says pairs of bronze pieces were joined along a seam and exposed to Nb vapor by sputter deposition during heating at about 715 C, so that the diffusion bond formed and Nb3Sn formed simultaneously, producing films that span the seam; magneto-optical imaging 'suggests' that supercurrent flows freely across the seam in several examples, and pre-coating the pieces before bonding gave varying seam coverage and no example comparable to the 'hot bronze' approach. The essay says the pieces are bonded first and Nb deposited and reacted afterwards, so that the brittle compound never experiences the bonding stresses; the abstract says the two steps happen together, and presents the supercurrent result as suggested by imaging, not demonstrated for every seam.
The “hot bronze” approach — bonding the substrate first, then coating — works better than pre-coating each piece and bonding afterward. The reason is microstructural: the Nb₃Sn layer forms after the mechanical joining is complete, so the brittle compound never experiences the stresses of the bonding process. The seam is invisible to the superconductor because the superconductor formed after the seam.
The practical implications extend to both magnets and superconducting RF cavities. For magnets, joining enables construction of complex geometries from segments. For RF cavities, Nb₃Sn coatings promise higher accelerating gradients than pure Nb, but cavity fabrication requires joining half-cells. A superconducting seam means the join does not limit performance.
The insight: do not try to preserve a fragile material through a brutal process. Instead, sequence the processes so the fragile material forms last, after all the brutal steps are done. The manufacturing order, not the material, is the engineering variable.