Lower Hybrid Current Drive at high density

The lower hybrid current drive programme in FTU has been devoted to provide experimental information on the issue of the current drive (LHCD) efficiency in plasma density regimes relevant for a reactor (i.e. line averaged density `ne~1*10exp20 m-3). The CD efficiency is defined as eCD=ILH*`ne*R/PLH [10exp20 m-2*A/W], where ILH is the LH driven current and R is the tokamak major radius. On Alcator C at high plasma densities, CD efficiency was found to be eCD~0.12 for toroidal magnetic field BT=10 T. This value has to be compared with that achieved at much lower densities on the largest tokamaks, as JET or JT-60 , where eCD~0.3 was measured at `ne=0.2*10exp20 m-3.

The results obtained on FTU, eCD~0.2 at density and magnetic fields close to those of Alcator C, show that the origin of such a difference can be traced back to the favourable scaling of eCD with <Te>, the volume averaged electron plasma temperature. Indeed, the main difference between FTU and Alcator C experiment is that considerably higher Te values are obtained in FTU, peak values being Te0> 2.0 keV against Te0~0.6 keV.

In FTU full CD regimes cannot be achieved with the present available power, but eCD can be also evaluated in the partial CD regimes, because the effect of the residual electric field on the suprathermal electron tail is within the experimental errors.

The experimental data are presented in Fig. 1, where ILH/Ip is plotted versus the parameter h = [ PLH/(ne Ip R) * 6/(Zeff+5)]. The linear extrapolation to ILH/Ip=1 yields the correct value for eCD, i.e. eCD~0.22 in full CD operation. In addition, it must be pointed out that the correction for Zeff is quite negligible at high densities, where Zeff~1.

Fig. 1 - Plot of the ratio of the LH driven current to the total current versus the quantity h defined in the text. Different symbols refer to different densities (see inset). Note how a fraction of about 50% has been driven at `ne~0.9*10exp20 m-3 and how the points at high density are indistinguishable from the low density ones.

For the data of Fig. 1 the accessibility to the FTU plasma core is deeper than 2/3 of the minor radius at first LH ray pass. Indeed a degraded efficiency is always observed as soon as the ray penetration is shallower.

The CD efficiency show no clear trend with `ne or any other macroscopic plasma quantity, except <Te>. The clear increase of eCD with <Te> is evidenced in Fig. 2, where the FTU eCD data, selected as the average at four values of <Te> inside the spanned range 0.36 - 0.92 keV, are compared with those from other tokamaks extrapolated to clean plasma conditions (Zeff=1). The largest <Te> interval available in the literature is examined, from the coldest (HT-6B) to the hottest (JET) device. It is evident that the favourable scaling with <Te> applies to all tokamaks.


Fig. 2 - Comparison of the highest values of the CD efficiency on various tokamaks as a function of the volume averaged electron temperature <Te>. For Alcator C the result for `ne=1*10exp20 m-3 is considered. The vertical bars indicate the range of results obtained for the given <Te>.

Several mechanisms for the enhancement of eCD with Te have been proposed by the theory , but it must be pointed out that Te affects also the PLH deposition profile and hence can change the fraction of the LH absorbed power as observed in JET . The beneficial effect of a Te increase on eCD has been observed also directly in FTU during a recent combined LHCD and electron cyclotron heating experiments . This is an encouraging result in view of the application of the LH waves on ITER like devices, since it confirms the possibility to attain efficiencies as high as those found at much lower density in JET and JT-60.

This work has been published in Phys. Rev Lett. V82, No.1 p.93 (1999)

Experimental reports