H-mode access and performance in the Mega-Amp Spherical Tokamak

Spontaneous transitions from the low “L-mode” to high “H-mode” of tokamak plasma confinement, first observed during neutral beam heating experiments on ASDEX, are now routinely achieved in many tokamak experiments. The H-mode regime is attractive as it offers the possibility of enhanced confinement, and thus a route towards a more “compact” and cost-efficient fusion power-plant. Transition to H-mode is now routinely achievable in the Mega-Amp Spherical Tokamak (MAST) [A. C. Darke et al., Fusion Technology 1994 (Elsevier, Amsterdam, 1995), Vol. 1, p. 799] for both Ohmically and neutral beam injection (NBI) heated plasmas (PNBI∼0.5–1.7 MW). H-mode plasmas can be either center stack limited or X-point diverted, exhibiting regular Type III edge localized modes (ELMs). Global confinement in H-mode with low frequency ELMs is consistent with the international IPB(y,2) scaling and exceeds the scaling by a factor ∼1.5–2.0 for high performance discharges. Confinement degrades with increasing ELM frequency (which in turn scales with power and density) as for conventional tokamaks. Densities above the Greenwald limit (G∼1) have been achieved for plasma currents up to 0.8 MA using gas-fueling, and up to 0.9 MA using a low field side multi-pellet injector. High field side fueling, on the other hand, can be supplied via a gas-feed located at the center-column mid-plane, this technique having been found to dramatically enhance H-mode accessibility and quality. When combined with Connected Double Null plasma topology, a significant reduction in Ohmic L–H power threshold can be achieved; as a result, power threshold data are now in broad agreement with a number of the latest scaling law predictions. Following the transition to H-mode, power crossing the inner separatrix remains low, resulting in a high recycling scrape-off layer (compared with partial detachment in L-mode). To date, with NBI power limited to 1.7 MW, H-mode MAST plasmas have shown no evidence of having approached a beta limit (βN∼4.5li). High performance H-mode discharges are at sufficient poloidal beta (βp), however, to enable the first studies of the Neoclassical Tearing Mode, the MHD instability responsible for limiting the achievable beta in conventional tokamaks.