Lower Hybrid Current Drive and Heating Experiments at High
The operation of Lower Hybrid (LH) radiofrequency
system in FTU (6 gyrotrons @ fLH=8 GHz, 2 antennas)
has reached the value of 2.2 MW corresponding to a net power density of 6.2
kW/cm2 on the waveguide mouths with an average reflection
coefficient of 10%. With this level of power, current drive (CD) studies are
pursued in a density range typical for a reactor plasmas, namely at line averaged
density ne ~11020 m-3.
We focussed on: 1) the study of the collisional
coupling between electrons and ions and of the CD efficiency, and 2) the way
to establish and sustain Internal Transport Barriers (ITB). For the first
point we worked at plasma current Ip=0.5 MA and toroidal magnetic field BT
~ 7.2 T, in order to have good accessibility of the LH wave in the plasma
core, whereas for the second point BT was in the range
useful for the electron cyclotron heating (ECH) power available in FTU (fECH=140
GHz, PECH up to 0.8 MW, BT=5.3
T for on axis ECH resonance). The discharges run for the e--i+ coupling study
exibit complete stabilization of sawtooth activity, with more than 75% of
the current driven by the LH wave and with an increase of electron temperature
in excess of 2 keV and of the neutron yield of one order of magnitude.
The results are summarised in Fig. 1 where the results for discharge #20026 (Ip= 0.5 MA, BT= 6 T) are shown. The line averaged density at its maximum is 1.01020 m-3 and the temporal behavior of the peak density is shown. The fraction of driven current is estimated from Vloop to be about 0.3 MA with an increase in Te0 from 1.8 to 3.8 KeV. The neutron yield increases by a factor 7, corresponding to a Ti raise from 1.2 to 1.55 keV. The sawtooth full stabilization is reached but an m=1 activity persists. The launched N|| spectrum is peaked at 1.82.
Fig .1 - Time evolution of the main plasma quantities
in a FTU high density LHCD discharge. From top to bottom are shown the time
traces of: a) the central plasma density, b) the loop voltage; c) the central
electron density; d) the central ion temperature; e) the neutron rate; f)
the coupled LH power.
Electron ITBs generally set as a steep gradient in the electron temperature, in a sawtooth free plasma when a weak or negative central magnetic shear (WS, NCS) is established . In order to realize and maintain in time a WS-NCS, not sawthoothing plasma with a high central electron temperature, it is desirable to break the "ohmic" link between the electron temperature and the current density profiles. A way to realize such a breaking is to drive a substantial fraction of the plasma current non-inductively, producing at the same time a WS-NCS discharge.
In FTU we have recently obtained wide electron ITB at density up to ne0~0.9 1020 m-3 (ne ~ 0.6 1020 m-3), by combining ECH and LHCD both in full and partial CD regimes. We have thus shown that operations near the ITER density and BT ranges do not prevent electron internal barriers to set in. The LH waves in FTU control the current density profile j(r), driving a large part of the plasma current (sometimes the whole) and heating the electrons, whereas the EC waves are used as a very localised electron heating source at the resonance radius. The EC power is used either to benefit from this improved confinement by heating inside the ITB, or to enhance the peripheral LH power deposition and current drive by setting the resonance radius off axis.
Two successful scenarios have been developed and studied up to now. In the first one, LH waves established full CD conditions and complete MHD stabilization, prior to the EC wave injection. The latter is launched during the current flattop, with the EC resonance located very close to the magnetic axis. In this way we got ITBs both at low density (ne = 0.3 1020 m-3) and high density (ne = 0.6 1020 m-3). High central electron temperatures (T e0>8 keV) were obtained at ne = 0.3 1020 m-3, BT = 5.3 T, Ip = 350 kA with PLH =0.6 MW, and PECH = 0.35 MW . The q profile was not measured, but transport simulations showed a magnetic shear reversal region at r/aĆ0.35. At the border of this region, a large gradient developed: LT-1= (dTe/dr)/Te=30 m-1, corresponding to R/LT= 28 (R is the major plasma radius).
At higher density, ne=0.61020 m-3 (`ne0=0.91020 m-3), Te0 (central electron temperature) =5.4 keV was achieved with PLH =1.7 MW, and PECH = 0.7 MW at BT=5.3 T, Ip=460 kA. The neutron flux also increased by a factor 2 from the ohmic to the LHCD phase and reached a factor 2.5 in the combined LH+ECH phase. According to the transport analysis, the WS/NCS region extended up to half radius, because of a broad LH power deposition profile. At the border of this region (r/a ~ 0.5) LT-1=20 m-1. The current was not fully driven by LHCD. The residual Vloop was ╝0.4 V (Fig. .2), corresponding to a residual ohmic power POH╝ 0.2 MW. As usual with central ECH heating in conditions close to full LHCD, the HXR profile emission remained unchanged during the whole heating phase, indicating a stationary current density profile.
Fig. 2 Time evolution of the main plasma quantities in a FTU ITB discharge with LHCD in the current ramp-up phase and ECH power off-axis (Ip=0.5 MA, BT=5.5 T). They are compared with a ohmic shot. From top to bottom are shown the time traces of: a) plasma current; b) line averaged density; c) central electron temperature, d) LH and ECH power.
In the second scenario both ECH and LHCD were
applied early in the discharge during the current ramp up phase (dIp/dt=2
MA/s), thus taking advantage from any pre-existing WS/NCS associated to initial
non relaxed j-profiles. The ECH was applied off axis, before LHCD (PECH╝0.3
MW, rdep/a=0.2) thus broadening the initial temperature
and possibly triggering an off-axis LHCD. Fig. 2 shows the time evolution
of the main plasma quantities. The LH power was injected in steps slightly
higher than 0.3 MW up to 1.7 MW in 100 ms, in order to compensate for the
increasing electron density. In this way an electron ITB (LT-1=20
m-1) was sustained longer than 0.2 sec (6-7 confinement
times) well inside the current flat top. In this phase the driven current
fraction was ILH/I p ~ 50%, the
central density raised up to ne0 ~ 0.81020
m-3, Te0 exceeded 11 keV, the
ITB footprint expanded from r/a~ 0.3 to ╝0.4, Ti0 increases
from 1 to 1.6 keV. The neutron yield also increased during the main ITB phase
and it was three times larger than in a reference ohmic discharge. The ITB
was terminated at t>0.34 sec by an m=1 MHD tearing mode, related to a change
in the current density profile after the ECH switch off.
The local transport analysis showed that, in the WS region, transport is indeed reduced during the main heating.