Inertial confinement fusion (ICF) aims to induce implosions of D–T pellets to obtain a extremely dense and hot plasma with lasers or heavy-ion beams. For heavy-ion fusion (HIF), recent research has focused on “liquid-protected” designs that allow highly compact target chambers. In the design of a reactor such as HYLIFE-II [Fus. Techol. 25 (1984); HYLIFE-II Progress Report, UCID-21816, 4.82-100], the liquid used is a molten salt made of F 10, Li 6, Li 7, Be 9 (called flibe). Flibe allows the final-focus magnets to be closer to the target, which helps to reduce the focus spot size and in turn the size of the driver, with a large reduction of the cost of HIF electricity. Consequently the superconducting coils of the magnets closer to the D–T neutron source will potentially suffer higher damage though they can stand only a certain amount of energy deposited before quenching. This work has been primarily focusing on verifying that total energy deposited by fusion neutrons and induced γ rays remain under such limit values and the final purpose is the optimization of the shielding of the magnetic lens system from the points of view of the geometrical configuration and of the physical nature of the materials adopted. The system is analyzed in terms of six geometrical models going from simplified up to much more realistic representations of a system of 192 beam lines, each focused by six magnets. A 3-D transport calculation of the radiation penetrating through ducts, that takes into account the complexity of the system, requires Monte Carlo methods. The technical nature of the design problem and the methodology followed were presented in a previous paper [Nucl. Instr. and Meth. A 464 (2001) 410] by summarizing briefly the results for the deposited energy distribution on the six focal magnets of a beam line. Now a comparison of the performances of the two codes TART98 [TART98: A Coupled Neutron-Photon 3-D Combinational Geometry Monte Carlo Transport Code, Lawrence Livermore National Laboratory, UCRL-ID-126455, Rev. 1, November, 1997] and MCNP4B [MCNP – A General Monte Carlo N-Particle Transport Code, Version 4B, La-12625-m, March 1997, Los Alamos National Laboratory] for two different configurations of the system is discussed, separating the n and γ contributions, in the light of the physical interpretation of the results in terms of first flight and of scattered neutron fluxes, of primary γ and of secondary γ generated by inelastically scattered or radiatively captured neutrons. The final conclusions indicate some guidelines and suggest possible improvements for the future neutronic shielding design for a HIF facility.
Read full abstract