Stewart J. Zweben

Former physicist (1984-2018)
Princeton Plasma Physics Laboratory

How to Contact Me:

Zweben photo
Last update
Jan. 30, 2024

Sections of this website:

1) Background on tokamak edge turbulence

2) NSTX edge turbulence imaging

3) C-Mod edge turbulence imaging

4) Other edge turbulence videos

5) Publications, talks, courses

6) History and opinions

1) Background on tokamak edge turbulence:

The goal of fusion research is to make a practical source of electricity from nuclear fusion reactions, similar to the reactions which generate power in the core of the sun. Controlled nuclear fusion requires heating hydrogen atoms to about 100,000,000 degrees Centigrade, which turns the hydrogen fuel into an ionized gas, or plasma.

The word tokamak is a Russian acronym for toroidal magnetic chamber. The hot plasma is confined by the magnetic field inside a donut-shaped metal vacuum chamber. The hottest plasma is on the long axis of the torus, while the edge plasma nearest the wall is a relatively cool 100,000-1,000,000 degrees Centigrade.

The tokamak edge plasma is usually highly turbulent, with density fluctuations of 10-100% over a broad frequency range 10-1000 kHz. These edge fluctuations are highly elongated along the magnetic field but have a short spatial scale of 1-10 cm across the magnetic field. The outward motion of these filamentary turbulent structures causes undesirable transport of heat and particles out of the plasma and onto the wall.

The videos below show the space vs. time structure of the edge turbulence in the plane perpendicular to the local magnetic field in the NSTX and Alcator C-Mod tokamaks, as measured using the Gas Puff Imaging (GPI) technique. The goal of these measurements is to understand the physics of the edge turbulence and its effect on edge plasma transport, based on comparisons with theory and simulation.

The GPI videos on this website were made in collaboration with R.J. Maqueda, B. Davis and D.P. Stotler of PPPL and J.L. Terry of MIT. The GPI diagnostic is described in the review paper: Review of Scientific Instruments, 88, 041101 (2017), which can be found in the list of papers in Sec. 5 below.

2) NSTX edge turbulence imaging:


The photo in the middle panel above shows the GPI diagnostic setup inside the NSTX vacuum vessel, a schematic cross-section of which is shown at the left. The red ellipse indicates a neutral hydrogen gas cloud which is puffed into the edge plasma through small holes in a manifold just above the outer midplane of the torus. This gas puff is imaged through an optical viewport about 1 meter below and to the right. This allows the edge turbulence to be viewed approximately along the direction of the local edge magnetic field line B, showing the cross-section of the turbulence filaments across B.

The image at the right is one single frame from a GPI movie taken at 400,000 frames/sec using a Phantom 710 camera outside the vacuum vessel. The camera looks through a Dalpha (656 nm) line filter and the image is presented in a false (red) color scale. The size of the imaging region is 24 cm radially (left-right) and 30 cm poloidally (up-down), corresponding to the GPI viewing region shown at the left. The magnetic field is into this image plane, with the magnetic separatrix shown by the dashed line and the RF limiter shadow by the dotted line. The shot time is shown in msec at the lower right.

The brightness of the GPI images depend on the local plasma density and the neutral gas density (which peaks near the separatrix). The nearly isolated bright filaments shown just outside the separatrix are called blobs. There is a simple theory which predicts that these blobs should move radial outward toward the wall at about 1 km/sec (about the speed of sound in air). Blobs normally cause most of the outward transport in this region.

These NSTX GPI videos were made in collaboration with R.J. Maqueda and B. Davis of PPPL.

Videos of NSTX Edge Turbulence:

3) C-Mod edge turbulence imaging:

photo of turbulence photo of turbulence

At the right are simultaneous single frames from movies from the midplane and X-region GPI cameras of C-Mod. The cameras view the HeI light from a local hdlium gas puff. The locations of these views are shown by the square boxes at the left, each of which covers a region of ~6 cm x 6 cm perpendicular to the local magnetic field. The sample movie frame at the left shows the X-region and outer midplane images at the same time during a shot. The solid lines are the separatrix and the dashed lines are flux surfaces at 1 cm and 2 cm into the SOL. The videos below were taken at 400,000 frames/sec using two Phantom 710 cameras. In both cases the data is normalized by dividing each frame by the time- averaged frame in that movie. Signal levels below some level were set to zero for clarity.

These GPI videos were made in collaboration with J.L. Terry of MIT.

Videos of C-Mod Edge Turbulence:

4) Other edge turbulence videos (not GPI):

Below are three older videos of tokamak edge turbulence made before the development of GPI.

The ASDEX video was made about 1982 by H. Niedermeyer and shows a wide angle view of filaments at the edge made using a high speed color film camera at about 7000 frames/sec. The left side shows a direct view of the plasma and the right side shows the edge of the plasma as seen using a mirror (see Goodall, J. Nucl. Materials 111 & 112 11 (1982)).

The Caltech video was made in 1984 from an 8x8 Langmuir probe array in the edge of the small Caltech tokamak. These probes show the structure and motion of the edge turbulence with about a 2 mm spatial resolution (see Zweben, Phys. Fluids 29, 974 (1985)).

The TFTR video was made in 1985 and shows edge turbulence at the inner (low major radius) wall of the circular TFTR plasma. The filaments are surprisingly distinct, despite the location in the good magnetic curvature region (see Zweben and Medley, Phys. Fluids B (10): 2058 (1989)). Both TFTR parts 1 and 2 start only after about 10 seconds of playing.

Other videos: