The preliminary design of a high temperature blanket and compact power conversion system for a D-T reactor utilizing a mercury Rankine cycle and nonequilibrium MHD disk generators is described. The CFAR cycle is described in detail and a blanket geometry is proposed which is designed to give good heat transfer and low pressure drops for the two-phase boiling mercury flow in the presence of strong magnetic fields. The blanket utilizes beryllium and flibe in the low-temperature breeding zone and lithium aluminate in beryllium oxide pebbles in the high-temperature breeding zone to obtain a tritium breeding ratio of 1.07. With careful choice of materials and geometry, it has been possibly to have 46% of the neutron and gamma heat deposited in the high-temperature pebble beds at the back of the blanket. The design utilizes the many advantages of liquid-metal Rankine cycles, including low pressures, mass flow rates and pumping power, and superheating along with internal regeneration to achieve good thermal efficiencies of about 35% using the microwave superheater and about 32% without the microwave superheater. The capital costs are expected to be much lower than past design concepts because the MHD power generators and the mercury condensers and heat exchangers can all be placed inside the reactor vault; preliminary cost estimates indicate a very attractive cost of electricity for this advanced concept.
A conceptual design study of a reversed field pinch fusion reactor