The Affordable Robust Compact (ARC) fusion reactor is expected to continuously generate 525 MW of power by fusion of deuterium and tritium (D&T) fuel in a tokamak heated to form a 200 million Kelvin superhot plasma. As helium is generated from the fusion reaction, it must be removed from the reactor to limit fuel dilution and helium radiation losses. To do so, the mixture of gases must be exhausted from the reactor and the helium must then be separated from the unspent fuel (deuterium and tritium) before the unspent fuel is reinjected into the reactor. Since there is no natural source of tritium, ARC is designed to breed tritium to ensure the system is self-sufficient in tritium. An additional tritium requirement is to minimize the tritium inventory to reduce the risk to public safety and because tritium inventory limits the rate at which fusion power plants can be deployed. Membrane systems are promising for meeting the gas separation needs of ARC because of the ease of operation, safety, efficiency, and amenability to compact integration with reduced tritium inventory.
This work is one of the 10 LIFT (Laboratory for Innovation in Fusion Technology) projects, sponsored by Eni and run by MIT, aimed at accelerating the development of ARC. It focuses on understanding the design of membrane separation systems toward the goal of developing advanced membranes and identifying system configurations for D&T and He separation from the ARC plasma exhaust stream. Material balance is established on the reactor to inform requirements for the D&T/He membrane separation system. The impact of using different fueling methods - namely, gas puffing and pellet injection - is studied. Membrane separation system designs for ARC are identified and evaluated to guide the selection of appropriate membrane configurations and to define the targets for the development of advanced membranes specifically for ARC's needs. The analysis suggests that less than 1 sq. meter of D&T-selective membrane area may be sufficient to separate He from the reactor exhaust, and that a membrane-based separation system will not contribute significantly to the tritium inventory. In addition, the analysis shows that the gas puffing method of fuel injection is favored due to its lower inventory even though it has low fueling efficiency. These results provide insight to guide the development of a compact system to remove the fusion-generated helium to maintain the correct composition of the plasma, recycle D&T, and minimize the tritium inventory.
