Search

In recent experiments at the HL-2A tokamak, dynamic features across the low–intermediate–high (L–I–H) confinement transition have been investigated in detail. Experimental evidence shows two types of opposite limit cycles (dubbed type-Y and type-J) between the radial electric field (Er) and turbulence evolution during the intermediate I-phase. Whereas for type-Y the turbulence grows prior to the change in Er, for type-J the oscillation in Er leads turbulence. It has been found that the type-Y usually appears first after an L–I transition, followed by type-J before the transition to the H-mode phase. Possible roles played by zonal flows and the enhanced pressure-gradient-induced flow shear in suppressing turbulence, respectively, in the type-Y and type-J periods have been identified. In addition, during the I-phase of the L–I–H discharges a kink-type MHD mode routinely occurs and crashes rapidly just prior to the I → H transition. The mode crash evokes substantial energy release from the core to plasma boundary and further increases the edge pressure gradient and Er shear, which eventually results in confinement improvement into the H-mode.

A long-lasting (for hundreds of milliseconds) m/n = 1 energetic particle mode driven by trapped fast ions, other than conventional fishbone bursts, is studied theoretically and in comparison with HL-2A experimental results. The mode can be observed in weak shear tokamak plasmas during neutral beam injection with a mostly steady amplitude envelope of long-lasting magnetic perturbation signals. The dispersion relation and radial structure of the mode are calculated with a weak shear q-profile. Both the m/n = 1/1 component and its higher frequency m/n = 2/2 harmonics are found to be unstable, in good agreement with experimental observations on HL-2A. On the other hand, due to the feature of weak magnetic shear, the mode is also significantly different from bursty fishbones, especially the mode structure, temporal behavior, instability threshold and growth rate dependence on the fast ion gradient. The nonlinear evolution of the mode and the comparison with fishbone bursts are also further investigated.

The synergy between electron cyclotron current drive (ECCD) and lower hybrid current drive (LHCD) is investigated numerically with the parameters of the HL-2A tokamak. Based on the understanding of the synergy mechanisms, a high current driven efficiency or a desired radial current profile can be achieved through properly matching the parameters of ECCD and LHCD due to the flexibility of ECCD. Meanwhile, it is found that the total current driven by the electron cyclotron wave (ECW) and the lower hybrid wave (LHW) simultaneously can be smaller than the sum of the currents driven by the ECW and LHW separately, when the power of the ECW is much larger than the LHW power. One of the reasons leading to this phenomenon (referred to as negative synergy in this context) is that fast current-carrying electrons tend to be trapped, when the perpendicular velocity driven by the ECW is large and the parallel velocity decided by the LHW is correspondingly small.

Measurements with a toroidally and poloidally displaced three-dimensional set of Langmuir probe arrays have revealed details of turbulence, geodesic acoustic modes (GAMs), zonal flows and their interactions in the edge region of HL-2A tokamak plasmas. The coexistence of intensive low frequency zonal flows (LFZFs) of f < 4 kHz and the GAMs of 7 kHz < fGAM < 20 kHz has been unambiguously demonstrated. The poloidal and toroidal symmetries of the flows, including the GAMs, are verified experimentally. In particular, the coherency of the flows over a large toroidal scale of 2100 mm at a magnetic flux surface is emphasized. The LFZF packets are shown to propagate outward and inward as equally likely events, whereas the predominantly outward propagation of the GAM packets is analyzed. The nonlinear three-wave coupling of the flows with ambient turbulence is shown with a bicoherency analysis and an envelope modulation of the latter by the former. The intensity of the LFZFs is observed to increase and decrease with increases in ECRH power (~300–700 kW) and safety factor q ~ (3.5–6.2), respectively, whereas the intensity of the GAMs increases with increases in both ECRH power and q.

Non-local transport phenomena induced by supersonic molecular beam injection (SMBI) have been observed for the first time in the HL-2A tokamak in low density discharges. For low densities (less than 2.0 × 1019 m−3), besides the similarities with the phenomena induced by other methods in various tokamaks, the results on HL-2A show some experimental progress in this effect: the duration of the core Te rise induced by SMBI in HL-2A could be prolonged by changing the conditions of SMBI; both the bolometer radiation and the Hα emission decrease when the non-local effect appears in low density. A repetitive non-local effect can be induced by periodical SMBs. Experimental and calculation results suggest that an electron transport barrier is formed at the position outside the q = 1 surface when the non-local phenomena appear.