Internal Transport Barriers studies

Robust electron (e-)-ITBs are obtained in FTU by the combined injection of the LH and EC waves during either the plateau or the ramp-up phase of the plasma current Ip. The LH waves have two main tasks: to control the current profile j(r) in order to built the e--ITB, by driving off-axis a large part of the plasma current, and to heat the electrons. The EC waves instead, are used as a localized electron heating source with two main purposes: to benefit from the improved confinement by heating inside the e--ITB (on-axis resonance), or to enhance the peripheral LH current drive by heating the plasma 4 cm off axis approximately. Ion transport barriers instead, have not been clearly individuated so far, because there is no ion (i+) heating source and the e--i+ collisional power (Pei) transfer is always too small.


Fig. 1. Time evolution for a steady ITB (Ip=360 kA, BT=5.3 T, EC resonance on axis) of: a) central and line averaged density; b) central electron temperature; c) max. of the normalized ion Larmor radius r*T,Max and normalized ITB size rITB/a; d) Zeff and peaking factors for Zeff and the bolometric losses e) applied RF power

In Fig. 1 it is shown the time evolution of an e-ITB, lasting longer than 37 times the global energy confinement time, E, and more than 1.5 times the current resistive diffusion time. The central plasma density, ne0, exceeds 1·1020 m-3 during the ITB (frame a)). The density increase during either the LH and ECH pulses is due only to enhanced recycling of neutrals at the walls. No sign of ne peaking is observed. While the central electron temperature increment is large, Te06 keV when ne0 is >1·1020 m-3, Fig.1 b), Ti0 rises only from 1 to 1.16 keV, because the e--i+ thermal equilibration time (eq0.18 s) much larger than E ( 0.02 s), makes Pei (0.12 MW) much smaller than the e- input power, PLH+ECH>2.2 MW. The full CD conditions, maintained in this phase assure the steadiness of the ITB size during the EC pulse, since there is no toroidal electric field to reshape j(r) upon a modification of Te(r), i.e. of the resistivity. In turn, the LH deposition profile is almost unaffected by a change of Te(r), according to the radial profile of the hard X-ray emitted by the LH generated fast e- tail. Transport code simulations show that the ITB footprint is very close to where the shear s of the safety factor q, s= r/q·dq/dr, starts to fall towards 0. In the ITB phase the effective plasma ion charge (Zeff) grows from 2 to 3, whereas its peaking factor increases only from 0.87 to 1.05, Fig. 1 d). Peaking factor is here defined as the ratio between the values averaged along a central chord and the innermost one of those viewing externally to the ITB (r/a=0.51 in this case). However, only half of Zeff follows from an increment of the impurity content, the other half comes from the higher average ionization level of Mo that is the first wall material, and consequently the dominant impurity in FTU. As Te0 raises from about 2 to 6 keV, <ZMo> grows from 30 to 37, implying Zeff0.5 for the measured Mo concentration in the range of 10-3 times ne .The rise of nZ is common in FTU upon an increase of the thermal load on the walls. The recent technique of coating the vessel walls with a boron film has reduced the high Z impurity influx with respect to the past, when only 1 MW of PLH produced Zeff1 at the density under consideration.

While the intensity and height of an ITB are controlled by the available power per particle, at the point that Te0>15 keV are attained when n e00.5·1020 m-3, the ITB radial size in FTU is controlled by the LH power deposition profile through q(a), the q value at the plasma edge. Figure 2 shows how the ITB expands as q(a) decreases. The barrier footprint moves outwards because the radius rs=0 where s becomes =0 is also shifted outwards by the more peripheral LH deposition. The data marked with full squares are obtained in over CD conditions, i.e. when a counter OH current in the plasma core broadens the final q(r) profile and moves outwards the point where rs=0.


Fig. 2. Variation of the ITB size normalized to the FTU minor radius a, versus q(a). The shrink of the ITB when increasing q(a) is due to the outward shift of the LH deposition. + symbols refer to Ip ramp-up phase, triangle to plateau, squares to discharges in over CD conditions

The plot in Fig. 3 of the increase of the ion thermal energy density Wi=(Wi|ITB-Wi|OH) versus the increment Pei=(Pei|ITB-Pei|OH) inside the ITB radius provides information on the global ion transport properties inside the ITB. The upper y-axis is labeled with the equivalent total power to ions inside half the radius. The good linear correlation between the two quantities proves the direct collisional heating of the ions inside the ITB and gives an incremental ion energy confinement time E,i-incr24.6 ms, larger than E, whose value is on average 20 ms. An ion heating efficiency of 3.6 keV/MW/1020 m-3, is derived from these data, larger than for electrons, for which we get about 2.2 in the same units.


Fig. 3. Behavior of the variation of the ion thermal energy density versus the collisional input power density, inside the ITB radius

The global transport properties of the ITB discharges are illustrated in Fig. 4, where r*T,Max (ratio of the ion Larmour radius to the plasma radius) is plotted versus the confinement enhancement factor (H97=E/E97, ratio of E to the ITER97-L scaling). Different time windows during OH, LH only, and LH+ECH phases are considered. The strongest ITBs, obtained with the combined action of the LH and EC waves, show the highest confinement. The clear increase of H97 with r*T,Max demonstrates the effect of the strength of the barrier on the global properties of the plasma. The departure of E from the L-scaling takes place very close to r*T = 0.014, assumed as the threshold for the onset of an ITB in JET. The transition from the normal to the improved confinement occurs smoothly, as it happens for the building of steep Te profile. No signature of sudden change of any property from non-ITB to ITB regime has so far been found in FTU.

Fig. 4 Global transport properties of the ITB discharges showing the improvement in confinement (by the factor H97, as soon as r*Tmax is over the threshold.

Experimental Reports