ALBERT

Notes: Abstract The injection of frozen pellets composed of the isotopes of hydrogen has become the leading candidate for refueling fusion power reactors based on the tokamak concept. This lofty position has been reached partly as a result of efforts to find an attractive solution to the perplexing problem of depositing atoms of fuel deep within the magnetically confined, hot plasma, and because of some recent experimental successes. To some extent, the relative merits of this technique will depend upon the distance that the cryogenic pellet will penetrate such a plasma, and the early exploratory research has addressed this problem on both theoretical and experimental fronts. The conclusion from the theoretical effort is that a protective blanket consisting of hydrogenic gas or cold plasma will envelope the pellet and partially shield the surface from the intense plasma heat flux. The blanket prolongs pellet lifetime, but penetration to the plasma center might require pellet injection velocities in excess of 10 km/s. The need for central penetration has not yet been established either theoretically or experimentally. The experiments performed to date have verified the existence of a shielding mechanism in general, and pellet ablation models that incorporate neutral gas shielding in particular are in adequate agreement with the experiments. Magnetic shielding effects are expected to contribute to, but not dominate, self-shielding in the higher plasma temperature regimes of the future. The tokamak plasma has demonstrated a surprising resilience even to massive density perturbations caused by the large refueling pellets used in present experiments. The characteristic discharge behavior is qualitatively not unlike that observed with gas puffing; but, for the first time, central plasma fueling has been studied, and this does not appear to be superior to refueling by partial pellet penetration. If relatively large pellets containing a significant fraction of the total plasma charge are acceptable in the present resistive plasma regimes, then it can be argued that they should have little impact on the gross stability of a hot thermonuclear tokamak plasma. Large pellets are preferable from the standpoint of attaining deep penetration, and this has important implications for the technology of pellet injection. The interesting velocity regime of 1 km/s has already been achieved with simple gun-type devices and this should be adequate for near-term tokamak experiments. Further improvements are anticipated, but the 10 km/s and above regime is uncertain; and, if current theory and experiments extrapolate to the future, such velocities might be desirable but unnecessary.