Superconductive thin films for RF applications

Abstract

Listen

Translate article

A particle accelerator is a complex machine, made of different parts constantly evolving in step with the technology advancements. One part of the accelerator, the LINAC, has nearly reached its maximum potential. This structure provides the acceleration through resonating cavities that under RF excitation transduce power into accelerating gradient. Superconductive structures are necessary to reduce the energy drain that these LINACs require. Niobium is the material of choice thanks to its high Tc and high field of first penetration Hc, but the research now has pushed the material to its limits. In addition, the infrastructure required to cool and operate a superconducting LINAC is very energy intensive and costly. New research avenues are opening aimed at reducing the costs of operating a superconducting LINAC. One of them is to create thin layers of niobium and niobium alloys on a high thermal conductive material, copper. If successful, the cost reduction in creating a functional Nb on Cu cavity versus a full Nb cavity will be substantial: a lower quantity of Nb will be required and better cooling will be achieved thanks to the higher thermal conductivity of Cu. This work aims at depositing Nb and Nb alloys on Cu substrates by using Chemical vapour deposition methods (CVD, PECVD, ALD) to verify if the obtained films are superconductive and can be employed in SRF applications. A series of different CVD chambers have been designed, built and evaluated to deposit the niobium-based superconducting films. Gas flow simulations were made by using Ansys Fluent to aid the understanding of the deposition results. The importance of a properly designed system has been highlighted, influencing factors like growth rate, uniformity and quality of the films. The deposition results were studied by electron microscopy (SEM, TEM, FIB), X-Ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS) to investigate their structural properties while the superconductivity was assessed by residual resistivity ratio (RRR) measurements, SQUID and localized magnetometry. The use of the chlorine-based precursors was anticipated to be detrimental to the stainless steel reactors used, due to the repeated cycles of hot HCl gas exposure followed by air / moisture. However, the heightened corrosion speed of the chamber resulted in the inclusion of iron precipitates in the niobium films. This issue was addressed by redesigning the deposition chamber. One of the thesis objectives was to evaluate the feasibility of using chemical vapour deposition to coat copper for SRF cavities. CVD processes based on the chlorinated precursors (NbCl5 and TiCl4) have been evaluated and successful depositions of Nb, NbN, and NbTiN on copper substrates has been demonstrated. Further, it has been demonstrated that Nb-based films can be deposited by CVD on copper with superconductive properties for the first time. It has been shown that CVD can be used to produce niobium coatings, with Tc matching the bulk material although with lower field of first penetration. The temperature influenced greatly the films: above 500 C no chlorine residue was present in the films, and no oxygen contamination thanks to the UHV deposition chambers. The best results were obtained by depositing at 700 C, with an RRR of 30 and large crystallites (500 nm across). The results suggest that the films would have more bulk Nb like properties if deposited at higher temperatures, but going above 700 C would compromise the Cu substrate by softening it or melting it. The NbN and NbTiN films possess structural properties matching the literature reference materials, however it was not possible to separate these films from the copper structures by dissolution. Consequently, their superconductive properties could not be measured, however the availability of new local magnetometry techniques will make this feasible in the future.