Abstract :
[en] This thesis is dedicated to the experimental study of methods to increase the shielded volume in passive magnetic shields made of high temperature superconductors (HTS).
Two main approaches are investigated. The first approach is to study how the size of the shield can be increased. The second is to study how the shielded volume can be extended in a shield of given size.
In the first approach, we study how large-scale or scalable magnetic shields could be obtained. Scalability means the possibility of building magnetic shields of various dimensions by keeping the same fabrication technique.
First, we study in detail the shielding properties and the behaviour of shields made as an assembly of superconducting coated conductor loops, obtained from second generation superconducting YBa2Cu3O7 coated conductor tapes that are commercially available. We show that such shields are an excellent alternative to bulk magnetic shields for shielding large volumes and fields of the order of 50 mT at 77 K. Different dimensions can be easily obtained by adapting the number and the length of the coated conductor tapes. This technique is shown to be very promising for building easily large size magnetic shields. However, unlike bulk magnetic shields, the shielding efficiency decreases rapidly when the applied field is no longer parallel to the main axis of the shield. The results in these configurations are analyzed to investigate the role played by the peculiar eye-shape cross-section of the shield on the various shielding current loops. We show that in the transverse field configuration, the structure can be used to locally modify the direction of the applied field. Finally, we take advantage of the scalability of such shields to build and study a triaxial structure that is able to attenuate a magnetic field of arbitrary orientation.
Next we study the superconducting properties and the shielding efficiency of YBa2Cu3O7 superconducting tubes made by two distinct fabrication techniques allowing large-scale or scalable magnetic shields to be obtained. The first technique is the electrophoretic deposition (EPD) which consists in the deposition of a superconducting thick-film (about 100 μm thick) on a metallic substrate. Compared to previous studies, our results show an improvement of the shielding properties and open encouraging prospects for a future development of magnetic shields made by EPD. For the second technique, the shield is obtained by the pulsed-laser deposition (PLD) of a superconducting thin layer (about 1 μm thick) on a textured metallic substrate. In the case where a continuous layer (i.e. joint-free) has been deposited successfully, we show that the persistent current flowing in the superconducting layer has a critical current density close to that of the commercial YBa2Cu3O7 coated conductor tapes, and point out the similarities with the coated conductor structures studied previously.
We also consider a similar material containing joints and compare their impact on the magnetic shielding in various configurations (DC/AC, axial/transverse). The results highlight the profound impact of nΩ level joint resistances on the superconducting shielding.
For the second approach used in this thesis, we study how to increase the shielded volume in HTS bulk magnetic shields of given dimensions. In the case of non-textured polycrystalline superconductors (Bi2Sr2Ca2Cu3O10 commercial tubes), we analyse the improvements resulting from closing one or both ends of a tube by a cap made of the same material. In this configuration the joint between the cap and the tube is non-superconducting. Numerical simulations are used to see how the volume can be further increased by increasing the thickness of the cap or increasing its critical current density. Then, we study the shielding efficiency of a melt-textured bulk YBa2Cu3O7 superconducting tube closed at one extremity with a cap in which the joint between the cap and the tube is superconducting. Our results show that such a tube can shield axial fields in excess of 1 tesla at 20 K and is therefore useful for shielding small volumes against high magnetic fields. The beneficial effect of a cap is also studied in the transverse field configuration. We show experimentally and with help of numerical simulations that, in this case, the superconducting character of the joint between the tube and the cap is mandatory.
In addition to the increase of the shielded volume, this thesis also addresses another problem relevant to future shielding applications. We investigate experimentally the effect of a magnetic field trapped initially in a bulk HTS hollow cylinder on its shielding performances. In the studied configurations, the initial applied field and the field to be shielded are mutually orthogonal. Remarkably the shielding properties are not affected provided the flux lines have not reached the inner hollow part of the cylinder. A trapped field inside the shield, however, is found to affect the threshold induction significantly. We then study how to efficiently demagnetize the superconducting tube without heating it above its critical temperature. By applying a single demagnetization cycle of adequate amplitude, we show that the initial shielding properties can be recovered.
