(FT/P2-4) Progress in Flibe Corrosion Study Toward Material Research Loop and Advanced Liquid Breeder Blanket

T. Nagasaka1), M. Kondo1), T. Muroga1), N. Noda1), A. Sagara1), O. Motojima1), A. Suzuki2), T. Terai3)
1) National Institute for Fusion Science, Toki, Japan
2) Nuclear Professional School, Graduate School of Engineering, The University of Tokyo, 2-22 Shirakata-Shirane, Tokai, Naka, Ibaraki 319-1188, Japan
3) Department of Nuclear Engineering and Management, Graduate School of Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan

Abstract.  Liquid breeder blanket is promising for DEMO and commercial fusion reactors. Flibe molten salt (BeF2 + LiF) is a very attractive liquid breeder material. Flibe blanket has been designed for Force-Free-like Helical Reactor (FFHR). One of the critical issues for Flibe blanket system is corrosion of the structural material, low activation ferritic steel, by HF (TF in tritium breeder). The reduction of HF by a REDOX (REDuction and OXidation control) reaction in Flibe has been demonstrated in Japan USA Program JUPITER-II. The next milestone to the blanket development is material corrosion research with non-isothermal convection loop (material research loop) to predict corrosion rate in the blanket condition. In the present study, development of purification technique for Flibe, its scale-up, and fundamental corrosion experiments have been performed toward construction of the material research loop and development of Flibe blanket. High purity Flibe is required for materials corrosion test, because HF, O and metal impurities heavily affect corrosion rate in Flibe. Purification was started with the volume of 50 g, and then it was increased to 150 g. In the present study, especially Fe impurity level was successfully reduced to 5-70 wppm, compared with a previous research (260 wppm). A blanket structural material (JLF-1 steel, Fe-9Cr-2W-0.1C) and a loop structural material (316L SS, Fe-18Cr-8Ni-2Mo) were exposed to the purified Flibe at 500 - 600oC for up to 2003 hr in static condition. The weight loss of JLF-1 at 550oC was equivalent to 0.83 micron/yr in corrosion rate. While, corrosion rate for 316L SS at 500 and 600oC was 3.3 and 5.4 micron/yr. The corrosion rates for both JLF-1 and 316L SS were considered as acceptable level for the structural materials. A material research loop was designed to investigate the effect of flow velocity (3 - 10 m/s) and temperature difference (100oC) in the blanket condition. From the above results, 316L SS is applicable to the structural material. 200 kg or more Flibe is required for the loop, and is feasible with the scale-up of the above-established process. Development of components of the loop, such as valve, flow meter and impurity censor, has been started. Feasibility of the material research loop was verified by the above results and other related R& D.

Full paper available (PDF)