Using Additive Manufacturing to Optimize FLiBe Coolant Blanket in Fusion Reactors

Vinny Fry

Duke University

Tuesday, September 5, 2017



PSFC Student Seminars

Fusion reactors have often been hailed as the holy grail of clean energy generation, though a power-generating reactor has never been built due to a multitude of limiting factors. One such factor is the immense 12-15 MW/m2 heat fluxes experienced by the inner wall of the reactor. Multiple groups have proposed the use of tungsten swirl tubes to withstand the heat generated within the reactor core. The primary focus of this investigation is to parameterize this ‘first wall’ interior structure to determine the highest achievable convective heat transfer coefficient given the many tungsten configurations enabled via additive manufacturing. Two general tube structures were considered: an orthogonal three-dimensional mesh of various diameters and spacings, as well as a swirl tube geometry with varying ‘tape’ thicknesses. The coolant liquid proposed is FLiBe (2LiF-BeF2) due to its high specific heat capacity as well as its ability to breed tritium, one of the fuels for the reactor. 

This was accomplished using theoretical calculations; computational fluid dynamics and conjugate heat transfer simulations in ANSYS Workbench; as well as an experimental setup to confirm tube pressure drop along the pipe. It was determined that heat transfer coefficients between upwards of 60,000 W/m2K were readily achievable, keeping the first wall temperature within the acceptable range at 1300 K. A multitude of designs proved to be feasible given the pumping power restrictions, though the suggested design going forward is a swirl tube with 2 mm ‘tape’ thickness and 2-3 m/s inlet velocity. Simulated pressure drop with water was accurate to within 30% of experimentally measured values, giving confidence in the credibility of the results.