The water-cooled lithium-lead breeding blanket is in the pre-conceptual design phase. It is a candidate option for European DEMO nuclear fusion reactor. This breeding blanket concept relies on the liquid lithium-lead as breeder-multiplier, pressurized water as coolant and EUROFER as structural material. Current design is based on DEMO 2017 specifications. Two separate water systems are in charge of cooling the first wall and the breeding zone: thermo-dynamic cycle is 295-328°C at 15.5 MPa. The breeder enters and exits from the breeding zone at 330°C. Cornerstones of the design are the single module segment approach and the water manifold between the breeding blanket box and the back supporting structure. This plate with a thickness of 100mm supports the breeding blanket and is attached to the vacuum vessel. It is in charge to withstand the loads due to normal operation and selected postulated initiating events. Rationale and progresses of the design are presented and substantiated by engineering evaluations and analyses. Water and lithium lead manifolds are designed and integrated with the two consistent primary heat transport systems, based on a reliable pressurized water reactor operating experience, and six lithium lead systems. Open issues, areas of research and development needs are finally pointed out.

The Water-Cooled Lead–Lithium Breeding Blanket (WCLL BB) is one of the two blanket concept candidates to become the driver blanket of the EU-DEMO reactor. The design was enacted with a holistic approach. The influence that neutronics, thermal-hydraulics (TH), thermo-mechanics (TM) and magneto-hydro-dynamics (MHD) may have on the design were considered at the same time. This new approach allowed for the design team to create a WCLL BB layout that is able to comply with different foreseen requirements in terms of integration, tritium self-sufficiency, and TH and TM needs. In this paper, the rationale behind the design choices and the main characteristics of the WCLL BB needed for the EU-DEMO are reported and discussed. Finally, the main achievements reached during the pre-conceptual design phase and some remaining open issues to be further investigated in the upcoming conceptual design phase are reported as well.

Four Nb 3 Sn conductor lengths were prepared according to the ITER TF conductor design and assembled into two SULTAN samples. The four lengths are not fully identical, with variations of the strand supplier, void fraction and twist pitch. Lower void fractions improve the strand support and increased twist pitches also lower the strand contact pressure but both tend to increase the AC loss and the lower void fraction also increases the pressure drop so that the mass flow rate in the strand bundle area of the cable is reduced.The assembly procedure of the two samples is described including the destructive investigation on a short conductor section to assess a possible perturbation of the cable-to-jacket slippage during the termination preparation.Based on the DC performance and AC loss results from the test in SULTAN, the impact of the void fraction and twist pitch variations is discussed in view of freezing the ITER conductor design and large series manufacture. A comparison with the former generation of conductors, using similar strands but based on the ITER Model Coil layout, is also carried out.The ITER specifications, in terms of current sharing temperature, are fulfilled by both samples, with outstanding results for the conductor with longer twist pitches.

No abstract

The full set of T CS measurements performed during 2000-2002 on conductor 1A of the ITER Central Solenoid Model Coil (CSMC) of the International Thermonuclear Experimental Reactor (ITER) is analysed with the extensively validated M&M code. Under the assumptions of uniform strand properties and uniform current distribution among strands, the performance of this ''average'' strand in the cable is deduced from the best fit of the measured voltage-inlet temperature characteristics. Two fitting parameters are chosen: an ad-hoc additional contribution Ôe extra Õ to the longitudinal strain of the average strand in the cable, and the average-strand (or cable ''effective'') index ÔnÕ of the electric field-current density characteristic. It is shown that the average strand in the CSMC performed less well than expected from the strand database--an increasingly negative e extra being needed to reproduce the coil behaviour at increasing transport currents, and that the cable effective n is clearly below the measured n of the strand. It is argued that this average-strand performance reduction in the CSMC is most likely related to mechanical load effects.