Plasma particle density profile versus collisionality in H-modes

n_peakingThe JET database on density peaking has been expanded with discharges dominated by radio-frequency heating. Regression analysis was performed on the expanded database, which, among several other parameters, takes account of neutral beam fuelling as a particle source. The scaling obtained in this way confirms that collisionality is the most important scaling parameter and that density profile peaking increases with decreasing collisionality in H-modes. A significant level of density peaking is predicted for ITER. The theoretical interpretation of the observed dependence of density peaking on collisionality is still controversial and poses a challenge to gyrokinetic models.

Poloidal rotation in discharges with ITB

Figure on the left shows a strong Internal Transport Barrier (ITB) expanding outwards in major radius, in JET Pulse No: 61324. From t = 4.04s onwards, the ITB is within view of the poloidal charge exchange diagnostic, which can measure poloidal rotation of carbon impurities in the plasma.

In the figure on the right the poloidal rotation of carbon measured in this discharge is shown at two times, one (red) when the diagnostic is viewing outside the ITB, and the other (blue) when the diagnostic is viewing the plasma both outside and inside the ITB. These measurements indicate there is a significant difference between the poloidal rotation either side of the ITB, with rotation in the enclosed region (R<3.4m) reaching much higher levels. Neoclassical theory predicts carbon poloidal rotation of a few km/s while the measured values are an order of magnitude larger within the ITB. Theoretical and modelling work is in progress to understand the cause of this rotation, and its link to ITBs. (Published in Phys. Rev. Lett.)

Study of ion loss during Edge Localised Modes (ELMs)

The JET Retarding Field Analyser (RFA) probe (photograph) provides the first direct measurement of ion energies convected outwards to the main chamber walls by Type I ELMs. The probe – inserted using a fast reciprocating drive into the scrape-off layer – is “bi-directional”, with sensors aligned perpendicular to the magnetic field facing the inner and outer divertors (the “e-side” and “i-side” marked in the figure). During ELMs, the i-side current density (jsat) dominates, showing that the ELM ejects particles preferentially in the outboard midplane region. Picturing the ELM as made up of a series of current “filaments” expelled simultaneously at multiple toroidal locations then propagating radially whilst losing energy rapidly along the magnetic field (as observed on MAST, ASDEX Upgrade and DIII-D) provides an explanation for the structure seen in the probe detector currents. The ion current measured inside the RFA (icoll) in the example shown here is due to ions which have been able to surmount a potential barrier of 400V. A new transient parallel loss model of ELM energy dissipation (published this January in Plasma Physics and Controlled Fusion journal) reproduces the measured i-side collector current (black line), showing that the ELM convects ions to the walls with energies characteristic of the H-mode pedestal region.

JET Restart Status

Since the start of Neutral Beam injection into plasma in January, operation of the upper half of the Neutral Beam system on Octant 4 has been strongly affected by the presence of an obstruction in the vicinity of the Neutral Beam Duct and High Vacuum Rotary Valve. After an intervention to remove the foreign body from the Octant 4 Neutral Beam Duct and to repair a water leak in an Octant 8 neutraliser, JET resumed plasma operation on 3 March. The vessel conditions are good, as shown by the fact that an H-mode was demonstrated already on 8 March with auxiliary heating power in the range of 8 ~ 10 MW, obtained during 5-6 second neutral beam pulses from Octant 4. Vessel conditioning and commissioning of all systems -including Ion Cyclotron Resonant Heating, Lower Hybrid Current Drive and diagnostics- is under way. The Experimental Campaigns are foreseen to start after Easter 2006.

Heat wave damping by Internal Transport Barrier

In experiments with modulated Ion Cyclotron Resonant Frequency (ICRF) power is deposited at two locations, at the centre (R~3m) and off-axis (R~3.6m), in a plasma with an Internal Transport Barrier (ITB) around R~3.35m. Two heat waves are generated and propagate towards the ITB. The amplitude (red squares) and phase lag (blue circles) of the resulting heat wave across the plasma are shown. The amplitude is damped strongly when it meets the ITB from either side. The phase rise shows that the wave slows down. This demonstrates that the ITB is a narrow layer with reduced heat diffusivity and with transport properties consistent with turbulence suppression. (Published in Phys. Rev. Lett.)

Distribution of fast ions with off-axis heating

Distribution of fast ions with off-axis heating
In experiments with off-axis Neutral Beam Injection, the radial distribution of fast injected tritium ions has been studied. This distribution is measured indirectly, by observing 14 MeV neutrons originating from fusion reactions between the fast tritium ions and the deuterium in the plasma. Plasmas with a high edge safety factor (q95 ~ 8.5) show a radial distribution of fast tritium ions in good agreement with simulations using the transport code TRANSP, peaking off-axis where the beams are aimed. However, at more ITER-relevant, low q95 ~ 3.3, some fast ions appear to be transported to the plasma centre, in disagreement with simulations. This anomalous behaviour does not appear to be related to sawteeth or MHD. Efforts to develop an understanding are ongoing. The results of this study could be important to improve predictions relating to the off-axis neutral beam current drive on ITER.

Material migration in outer divertor

Material migration in outer divertor
Previous experiments on JET (2001) showed that carbon injected at the top of the vacuum vessel was deposited mostly on the inner divertor. This was understood as a result of Scrape-Off Layer (SOL) flows from the outer divertor to the inner divertor, which carried the injected carbon with them. The observation suggested most material released from the first wall would be deposited on the inner divertor. More recent experiments with carbon (13C) injection in the outer divertor region rather than from the top of the vacuum vessel have shown that material migration is more complicated in this region. Part of the injected carbon is deposited locally onto the outer divertor tiles close to the injection point, and a significant amount is transported to the inner divertor. The injected carbon is also found on a deposition probe on top of the machine, indicating strongly that it migrated up to the X point, through the confined plasma, back into the SOL and then down to the inner divertor leg. Modelling is underway to understand how the observed migration arises, and to improve predictions for ITER, where tritium accumulation by co-deposition must be strictly controlled.