The Manufacture of SiC for the Application to Joining of Nuclear Fuel Cladding

Abstract

With the recent surge of interest into Accident Tolerant Fuel concepts and High Temperature Reactors, there has been rapid progress in the design and manufacture of nuclear-grade SiC-SiC composite components. In particular, composite fuel cladding for current generation Light Water Reactors is under development due to the superior oxidation and high temperature behaviour of SiC. A significant manufacturing challenge for this technology is the joining of fuel end caps to the cladding tubes after fuel insertion. With no liquid phase, SiC cannot be welded in the same way as conventional metallic cladding has been. The development of a suitable joining method has been focused on achieving hermetic seals, with the irradiation stability and corrosion resistance of the rest of the cladding component. The majority of potential joining methods are high energy processes with estimated processing times that can be measured in days per fuel tube. It is therefore highly desirable for faster methods to be developed that can be economically comparable to the higher turnover of zirconium cladding manufacture. This project focuses on two SiC manufacturing methods that have the potential to be applied successfully and economically to SiC joining: Laser Chemical Vapour Deposition (LCVD) and Flash Spark Plasma Sintering (F-SPS). An LCVD method using Dichloromethylsilane (DCMS) as a precursor was successfully developed and the data obtained could be compared to literature data for conventional Chemical Vapour Deposition (CVD). Although it was predicted that the use of laser heating and DCMS would improve the rate at which material is deposited compared to conventional CVD, the data from this investigation does not support this. It is believed that with further optimisation of the LCVD process, deposition rates can be improved. It was also found that the SiC deposited from the DCMS precursor was more homogeneous and stoichiometric than the material deposited from Tetramethylsilane (TMS), a more commonly used precursor. Significant improvements were made to the LCVD method by using an annular 18 beam as opposed to a Gaussian beam, which improved the temperature distribution across the substrate. Additionally, the use of a continuous wave laser rather than a pulsed laser was found to produce more homogeneous SiC films. The project was also successful in the design and implementation of a novel FSPS SiC manufacturing process that does not have a pre-pressing step. The maximum density achieved from this process is 77% of the theoretical density and a complete transition to a SiC was observed, the less irradiation-stable phase. Although the F-SPS process is significantly quicker than conventional SPS, the SiC produced is not suitable for the nuclear fuel cladding application and further process optimisation is required. Other SPS investigations have shown that the degradation of fibres within composites exposed to the SPS environment is a significant barrier to producing high quality SiC composite joints. In all this project concludes that the LCVD method has more potential for the application to the joining of SiC-SiC composite cladding, since it produces higher quality, stoichiometric and cubic-phase SiC. It is recommended that further work is undertaken in improving the deposition rate in order to improve the feasibility for LCVD to be used as a mass-scale manufacturing method. Whilst SPS and F-SPS are much faster manufacturing methods, the challenge of retaining cubic phase SiC that is also dense and therefore hermetic may be too great. Both methods are estimated to produce SiC joints in less time than other common manufacturing methods and should therefore be subject to further investigation and development.

Date of Award	1 Aug 2019
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Ping Xiao (Supervisor) & Timothy Abram (Supervisor)