Magnetic confinement of thermonuclear plasmas

Toroidal reactors will require a ∼10‐keV DT plasma with nτ (density times confinement time) exceeding ∼1014cm−3sec. Present‐day toroidal plasmas come within 2–3 orders of magnitude in respect to nτ, and within less than an order of magnitude in respect to temperature. Toroidal loss processes appear to be diffusive, so that nτ increases with size. In the low‐pressure ("low‐β") approach, typified by the tokamak, the principal physics problems are: (1) to develop scaling laws for plasma diffusion (a problem of nonlinear microscopic instability); (2) to minimize impurity evolution from plasma‐wall interactions (a surface physics problem); (3) to find grossly stable configurations maximizing ntgr; (a MHD‐stability problem). The high‐β approach, typified by the toroidal θ‐pinch, is related to item (3). Plasma pressure and density are raised to their limits by special MHD‐stabilization techniques, including feedback. Development of shock and compressional heating techniques plays an important role. An alternative reactor approach is the open‐ended magnetic bottle, or "mirror machine." A non‐Maxwellian (beaminjected) ion population in the energy range >100 keV interacts with itself or with a plasma target. The nτ‐value depends on velocity‐space distribution rather than on size; present‐day nτ‐values need to be improved by 2–3 orders of magnitude. The principal physics problem is anomalous endloss — a problem of nonlinear high‐frequency instability, somewhat like laser action.