Modelling, Development and Testing of a Highly Compact, High-Field (RE)BCO Magnet

The strongest sustained direct current magnets in the world utilise the Bitter electromagnet architecture, which consists of a stack of slit Cu-alloy discs and insulation layers arranged to form a helical conduction path. However, resistive Bitter magnets require substantial energy to power and cool the coils, limiting their application to dedicated facilities. This thesis proposes, simulates, and experimentally demonstrates a new coil design based on the Bitter architecture, but employing high-temperature superconducting (HTS) rare-earth barium cuprate ((RE)BCO) bulk superconductors instead of resistive copper alloys. The (RE)BCO design has voltage, space, and power requirements that are orders of magnitude lower than those in an equivalent resistive Cu-alloy Bitter magnet, making it feasible to power with a low-power laboratory supply.

Accurately modelling the electromagnetic behaviour of (RE)BCO superconductors requires experimentally-measured data for the material’s thermo-magneto-angular anisotropy of both the critical current density and flux-flow exponent, Jc(T, B, θ) and n(T, B, θ), respectively. Whilst there are many examples of Jc(T, B, θ) and n(T, B, θ) published for (RE)BCO coated-conductors in the literature, such data for (RE)BCO bulks is scarce and incomplete. Therefore simplifying assumptions, such as assuming a constant Jc and n-value, are common when modelling HTS bulks. These assumptions can lead to significantly different current distributions within a superconductor under an applied magnetic field. This is illustrated in this thesis using an experimentally-validated two-dimensional (2D) H-formulation finite-element model of a permanent magnet approaching an HTS coated-conductor tape, for which Jc(T, B, θ) and n(T, B, θ) are known.

In order to address this lack of data, a new sample preparation method was developed, enabling the characterisation of bulk superconductors using a four-point transport measurement. Extensive Jc(T, B, θ) and n(T, B, θ) data were measured for top-seeded melt growth (TSMG) GdBCO-Ag as well as for both TSMG and single-direction melt growth (SDMG) EuBCO-Ag bulks for the first time. The results were compared to Jc(T, B⊥) data measured using traditional magnetisation characterisation methods, with both techniques showing good agreement.

The measured Jc(T, B, θ) and n(T, B, θ) data were incorporated into a 2D axisymmetric H-formulation model of the Bitter-like HTS magnet to simulate and understand its electromagnetic behaviour. Phenomenological differences between Cu and HTS coils were investigated and explained, and the effects of varying key parameters in an HTS coil were explored. Different coil configurations were studied, which incorporated a bulk Bitter-like coil into a multi-coil magnet.

The practical formation of a Bitter-like coil with a helical current path from HTS bulk wafers requires the formation of mechanically robust conducting joints. Two different jointing methods were experimentally investigated. The effect of changing the heating profile on the resulting joint resistances was investigated, and results for both superconducting and resistive joints are shown in this thesis. Six multi-layer prototype coils were then manufactured from commercially available TSMG GdBCO-Ag bulks, using the silver-diffusion jointing method. The coils were tested, achieving a maximum field of 0.49 T at 960 A at the centre of the 6 mm diameter bore of a 20-layer coil at 77 K.

The Bitter-like HTS bulk electromagnets introduced in this thesis represent a promising novel coil architecture that could enable the realisation of magnetic fields exceeding 10 T with low material, cooling, and operating costs. This work demonstrates a clear pathway to the fabrication of highly compact high-field systems. In addition, the characterisation data for the (RE)BCO bulks will be published, providing the global bulk modelling community with access to an extensive Jc(T, B, θ) and n(T, B, θ) dataset for the first time.