[Better copy in Postscript L2: Complete, or p1 p2 p3 p4]
The H-1 heliac is the centrepiece of a toroidal plasma physics program at the Australian National University. Experimental results will be summarised; diagnostics are described in separate papers. Computational MHD and transport studies, and theoretical work on ballooning modes and radio frequency wave propagation will be reviewed.
The H-1 heliac [HB90,BB94] is a medium-sized (R = 1.0m, ; <!!a!!> 0.2 m) three-field-period helical axis stellarator designed for basic studies of "current-free" toroidal confinement in both quasi-continuous operation (at B < 0.25 T) and in pulsed fields (B <1 T). Controlling the ratios of currents in its five coil sets allows a very wide range of low-shear, high-transform-per-period, closed flux configurations to be accessed e.g. with 0.6 < < 2 , and with full-width magnetic wells of both signs (+5% to -2 %).
H-1 was operated in quasi-continuous mode for the past year; the magnetic field is applied continuously at reduced strength, and the rf power is applied whenever required, typically in long pulses 100 every 10-20 seconds, or for shorter pulses at higher rates. This completely avoids transient magnetic field effects on the configuration, simplifying magnetic diagnostics, and is a very convenient mode for performing basic plasma physics experiments in argon, helium and hydrogen. Experimental parameters are somewhat reduced, but a wide range of parameters were obtained from helicon wave rf produced [BB95] plasma (1e10 < [[onesuperior]].< 5e12 3,; 3< < 50 ). Regimes of interest include a finite b. regime (0.5%,; B0 0.1) , and a low to medium collisionality regime ([[onesuperior]].1012 -3,; B0 0.2) . Finite and low collisionality are simultaneously possible in helium and hydrogen. A range of advanced tomographic/mm-wave diagnostics is described in a separate paper.
In hydrogen and helium, higher densities and temperatures, and lower ionization outside the plasma volume have been achieved by a high throughput (300 Torr.l/s), directed gas puffing system. For example, a 10ms, 80 Torr.l/s pulse in helium provides a peaked profile ([[onesuperior]].1e12 ,; 20-30 ) at B0 = 0.13 , for which conditions a lower puff rate or static fill produces a flat or hollow profile of lower density. Data, including profile evolution, is presented in fig. 2.
High b. is obtained in helium (higher ) and argon (higher [[onesuperior]]. ). Operation in the lighter gases is preferred, but argon plasma is more tolerant of the simultaneous insertion of large numbers of probes.
Results are in broad agreement with VMEC predictions of pressure-driven currents, (e.g. Pfirsch-Schluter currents) (fig. 1). Under some conditions, a net current of similar magnitude and uncertain origin is observed, and can markedly change the baseline of data such as fig. 1b.
The radial profile of plasma density tends to be gaussian-shaped about the magnetic axis at low fields, and tends to sharpen to a slightly rectangular shape at higher fields in helium and argon. Two profiles are shown in fig. 2, to illustrate this, under different vertical field conditions.
Energy confinement scaling studies were performed to investigate the effect of magnetic configuration and of radio-frequency plasma formation and heating, and to compare with neoclassical scaling. Plasma energy was estimated from data obtained from three diagmagnetic loops at different cross-sections, and corroborated by Langmuir probe measurements of and , and some spectroscopic temperature estimates. Although initial temperature measurements are not in close agreement (spectroscopic estimates in helium are 2-3 times lower than probe estimates), measurements of the ion saturation current profile, which differ is scaling from energy measurements only by a factor of T are in reasonable agreement with energy measurements from the diamagnetic data.
Initial results are complicated by the scaling of the helicon plasma generation method in which electron density is found (in smaller machines [LB95] to scale linearly with magnetic field. At very low fields, confinement is dominated by atomic physics. However, the plasma kinetic energy is found to increase with a power of B0 between 1 and 2, until a limit is reached when the (low) applied rf power can no longer sustain the plasma. This saturation or decrease occurs for argon only at low power, and for powers of 60kW (the maximum routinely applied in these experiments) in helium and hydrogen for magnetic fields above 0.1 and 0.05 Tesla respectively. Figure 3a shows and example of the saturation and ultimate decrease in Wperp in hydrogen, in contrast with the steady increase for argon.
[Click]
Depending on details of mode structure and boundary conditions (currently
under investigation), it is likely that in hydrogen at least, the saturation is
simply caused by insufficient density for fast wave propagation at the higher
magnetic field. It is expected that this will be overcome by increasing rf
power, as was the case in argon, or by increasing the applied frequency.
Initial results of energy confinement measurement from plasma energy decay
rates after the rf heating in removed, indicate that local confinement
measurements may be feasible, although conditions are likely to be non-linear.
The floating potential varies from near zero to markedly negative ( 200 ) for a range of conditions. A typical profile is shown in fig. 4. The plasma potential is estimated by correcting for the plasma temperature, within the uncertainties discussed above. The floating potential is affected strongly by rf, and by gas puffing. The magnitude of the potential falls rapidly when rf power is removed, or a time scale of a few ms when gas is puffed. The usual sign of the H-1 potential (negative) is of opposite sign to that expected by a simple model of plasma biasing by rf applied to an "electrode" in the plasma (in this case, the unshielded antenna, centre-earthed antenna).
A low frequency oscillation in the kHz range is often observed in Ar plasma, (and sometimes in Helium, in the tens of kHZ range,) bearing some similarity to drift wave modes observed in SHEILA[SB89]. The mode structure is low order (see fig. 4b), and there appears to be correlation with pressure gradients, or plasma currents [SB95].
The effect of the sign of the shear on drift wave stability is also of interest [NJMP95]. Drift wave stability is being studied in the ballooning representation using both the cold-ion and gyrokinetic models. Visualization of global and local stability criteria is being developed into a valuable tool for searching for and identifying instabilities both in the experiment, and in numerical simulations.
A Monte Carlo transport code using VMEC equilibrium information is being written to iteratively determine self-consistent electric fields and transport in the H-1 plasma. In a separate study [GG95], optimization of high b. equilibria, transport and bootstrap current has led to an economical compact reactor configuration, the three-field period modular heliac-like helias (MHH3). The effect of rf and fluctuations on transport and current drive is also being investigated [PD95,PB95].
[SB89] SHI, X-H., BLACKWELL, B. D. and HAMBERGER, S. M. Plasma Physics and Controlled Fusion. 31, 13 (1989)
[SB90] S. M. Hamberger, B. D. Blackwell, L. E. Sharp and D. B. Shenton, Fusion Technol. 17 123 (1990).
[BB94] B. D. Blackwell, G. G. Borg et al (Proc. 15th Int. Conf. Seville, Spain, 1994), IAEA-CN-60-60/A6/C-P-4.
[DHP94] R. L. Dewar, S. R. Hudson and P. F. Price Phys. Letters A 194, 49 (1994).
[BB95] G. G. Borg, B. D. Blackwell, S. M. Hamberger, D. L. Rudakov, D. A. Schneider, L. E. Sharp, M. G. Shats and B. C. Zhang, Fusion Eng. and Design, 26 191 (1995).
[DDGH95] M. G. Davidson, R. L. Dewar, H. J. Gardner and J. Howard, Aust. J. Phys. in press (1995).
[GG95] P. R. Garabedian and H. J. Gardner, Phys. Plasmas 2, 752 (1995).
[HSG95] T. Hayashi, T. Sato, H. J. Gardner, J. D. Meiss, Phys. Plasmas 2, pp752-759, (1995)
[LB95] P. K. Loewenhardt, B. D. Blackwell and S. M. Hamberger, Plasma Physics and Controlled Fusion, 37 229 (1995)
[NJMP95] H. Nordman, A Jarmén, P. Malinov and M. Persson, Phys. Plasmas in press (1995).
[PD95] V. Petrzilka and R. L. Dewar, Austr. J. Phys. in press (1995).
[PB95] V. Petrzilka and R. Balescu, submitted to J. Plasma Phys. (1995).
[SB95] "Magnetic Configuration Scans in H-1 Heliac" ,M. G. Shats, B. D. Blackwell, G. G. Borg, S. M. Hamberger, J. Howard, D. L. Rudakov and L. E. Sharp, Fusion Technology, 1995, accepted.