Convective Transport Suppression in the Scrape-Off Layer Using Ion Cyclotron Resonance Heating on the ASDEX Upgrade Tokamak

Turbulence properties in the scrape-off layer (SOL) in the presence of ion cyclotron frequency heating (ICRH) are compared to instances where it is absent. The discharges are all in a high-confinement mode ($H$-mode) regime. During ICRH, the SOL plasma density increases whereas turbulence large-scale and convective structures are shown to be suppressed. The probability distribution function is thus recorded to be closer to a Gaussian, and a net decrease in the low-frequency density fluctuations is reflected in the power spectra. Consequently, the level of turbulent fluctuations decreases significantly. Turbulence suppression is also reported during edge localized modes (ELMs) where both the ELMs-induced transport and duration are strongly affected. The increase of neutrals by gas puffing did not alter this behavior. We deduce that ICRH can be used as to suppress convective transport and reduce the ELM’s amplitude.

Figure 1

Solid lines and dashed lines indicate data taken in 21408 and 21409, respectively. Panel (a) illustrates the plasma current (in MA) and the normalized plasma density; the dip in the core plasma density during ICRH is zoomed on in (e). In (b) we plot the NBI and the coupled ICRH (in MW) power as a function of time. The $D_{\alpha }$ in (c) shows the ELMs, indicating that the discharge remains in $H$ mode. The neutral gas pressure at the divertor and the midplane (multiplied by 10) are shown in (d) where the increase at 5.5 s in 21409 is caused by the midplane valve’s opening shown in (g). The solid lines and the dotted lines in (g) indicate the injected gas density in 21408 and 21409, respectively, from the divertor valves. The dashed line illustrates the gas density injected by midplane valves. The boron II emission line from the antenna shield is plotted in (f) in $\mathrm{\text{photons}}/({\mathrm{m}}^2\mathrm{sr}\mathrm{s})$.

Figure 2

Panels (a1) and (a2) illustrate the average density profile from LiB for shots 21408 and 21409. Dashed lines and solid lines illustrate, respectively, data with and without ICRH. In  (b) and (c), $I_{\mathrm{sat}}$ is shown for a midplane and a high-field side target probe during 21409.

Figure 3

In (a), we plot $I_{\mathrm{sat}}$ (in A) of the midplane filament probe with [black (blue)] and without [gray (red)] ICRH. The initial times for the two plots are, respectively, $t_0\sim 2$ and 6.1 s. The empty time lapses are mainly caused by ELMs and negative biasing to avoid sparks that we have removed. In (b), we removed the empty time lapses in order to clearly show the radical changes occurring to $I_{\mathrm{sat}}$ when ICRH is switched on. In (c), we inserted $I_{\mathrm{sat}}$ from shot 20964 where $P_{\mathrm{tot}}=5.8\mathrm{MW}$ coming from 2.5 of NBI and 2.8 of ICRH. The fluctuations can thus be compared to 21408 at $t\sim 2\mathrm{s}$ with the same $P_{\mathrm{tot}}$ coming from NBI alone. In (d), the $I_{\mathrm{sat}}$ taken from discharge 21868 with $P_{\mathrm{tot}}=P_{\mathrm{NBI}}=7.5\mathrm{MW}$ is plotted. Another comparison is thus possible between the fluctuations shown in (d) and those in 21408 around $t\sim 6.1\mathrm{s}$ with about the same total power.

Figure 4

In (a), the PDF of $I$ is plotted in a semilogarithmic frame with dashed lines and solid lines for data with and without ICRH, respectively. In (b) and (c), we illustrate the power spectrum of $I$ as a function of the frequency $f$, and the autoconditional average (CA) as a function of time $\tau$ with the same legend for the type of lines as in (a).

Figure 5

In (a) we plot $I_{\mathrm{sat}}$ during 21408 where two ELMs are visible, one with an initial time $t_0=2\mathrm{s}$ (no ICRH) and the other at $t_0=6.1\mathrm{s}$ with ICRH. In (b) the autoconditional average of ELMs is illustrated for 21408 and 21409 and with and without ICRH. In (c) the two bar charts reflect the PDF of the ELM frequency $f_{\mathrm{ELM}}$ for 21408 with (filled) and without (empty) ICRH power. The stairs plot shows the same for 21049 with the presence of ICRH.