Buoyancy-Driven Two-Phase Flows of Liquid Metal contributing to the Generation of Electricity from a Fusion Reactor by Magnetohydrodynamic Energy Conversion.

Abstract

Fusion is desirable for providing the world’s future base-load power capacity due to its lack of greenhouse gas emissions, low environmental and safety risk, and large, secure fuel reserves. Fusion power plants are not expected to deliver electricity commercially until 2050, when it is expected that most fossil-fuelled power plants will have been removed from the global electricity generating mix. However, for fusion power plants to remain competitive with other forms of primary energy sources, the unit cost of electricity should be kept as low as possible. The objective of this work has therefore been to examine and compare a range of conversion methods for the purpose of increasing overall fusion plant efficiency. Model C of the Power Plant Conceptual Study provided a benchmark against which a proposed power cycle could be compared. The power cycle associated with model C, using a thermal input of 3991 MWt and neglecting cycle pressure losses, delivers 1780 MWe with a cycle efficiency of 44.6%. Allowing for blanket gain, but after deducting plant power requirements, an overall plant efficiency of 45.9% is achieved. The proposed power cycle uses a primary Brayton cycle, which takes helium directly from the blanket, to deliver 1688 MWe. A topping cycle, which uses two-phase liquid metal magnetohydrodynamic energy conversion, delivers 22 MWe. Three water-based bottoming Rankine cycles are used to extract heat rejected by the Brayton cycle to deliver 409 MWe. Small-scale experimental work was undertaken to verify the performance of the topping cycle and observe the circulation of a dense liquid metal by using buoyancy, thereby removing the need to pump liquid metal through a blanket. The proposed power cycle, without cycle pressure losses, delivers a total of 2119 MWe with a combined cycle efficiency of 53.1%. An overall plant efficiency of 56 % is achieved, an improvement of 10.1% over PPCS model C.