Plasma behaviour with molecular beam injection in the HL-1M tokamak

A new method of gas fuelling has been introduced in the HL-1M tokamak. The method consists of a pulsed high speed molecular beam formed by a Laval type nozzle. The velocity of the well collimated hydrogen beam is about 500 m/s. About 6 × 1019 molecules pass through the nozzle and into the vacuum chamber in each pulse. A series of helium pulses was injected into the HL-1M low density (e = 4 × 1018 m-3) hydrogen plasma. With penetration depth up to 12 cm, the ramp-up rate of the electron density, de/dt, was as high as 3.1 × 1020 m-3s-1 at steady state, and the resulting plasma density reached e = 5.6 × 1019 m-3. The profile peaking factor of the electron density, Qn = ne(0)/ne of about 100 ms after helium molecular beam injection (MBI) reached a maximum value of more than 1.51. The energy confinement time τE measured by diamagnetism is 26 ms, which is over 30% longer than that of the gas puffing (GP) results under the same operational conditions. The improvement of τE and increase of Qn for MBI were comparable to those of small pellet injection (PI) in HL-1M, as well as those of slow PI in ASDEX (Kaufmann, M., et al., Nucl. Fusion 28 (1988) 827). It is argued that the peaked density profile induced by the deepened particle injection is a factor essential for the confinement improvement apart from the isotope effect of helium particles, because the density peaking factor Qn is normally less than 1.4 for GP plasma in HL-1M. The particle confinement time with MBI increased sixfold in comparison with that before injection.