A comparison of experimental measurements and code results to determine flows in the JET SOL

Abstract

Two reciprocating probe systems, in the same poloidal position at the top of the JET torus but toroidally separated by 180 o , have been used to measure parallel flow in the Scrape-Off Layer (SOL). One system uses the entrance slit plates of a Retarding Field Analyser (RFA) to record upstream and downstream flux densities, and the second system uses two pins of a nine pin Turbulent Transport Probe, (TTP). Measurements have been made in both forward and reverse field directions.Results from both systems are similar.In the forward field direction, that is with B r ∇ drifts downward towards the divertor, there is a strong parallel flow in the direction outer to inner divertor. The flow generally has a low value, Mach number M~0.2, close to the separatrix and rises in the region of high magnetic shear close to the separatrix to a maximum of M~0.5 some 20mm outside of the separatrix. The flow in the reverse field direction is small, close to zero, and generally is again in the same direction as that for forward field close to the separatrix, with M~0.2. Code results using EDGE2D with drifts suggested an almost symmetrical flow about zero when the field direction was changed in an earlier work, [1]. However, this was for particularly low density, high temperature edge conditions, and the predicted symmetry is not evident for more usual edge conditions, reported here. Experimentally, the flow is found to be quite asymmetric about zero, particularly at high density. There is some symmetry in flow, but about an offset value of M~0.2. The form of M(r) is similar to experiment but the major code result is the low value of M generally < 0.1. The effect of gas (deuterium) and impurity (carbon) puffing in the code has been investigated. We are unable to say why the magnitude of M(r) from experimental data and codes do not agree. However, results presented in this paper suggest that the probe itself may be exhibiting an influence on the magnitude of the flow as measured in the SOL.

Outer divertor of ASDEX Upgrade in low-density L-mode discharges in forward and reversed magnetic field: I. Comparison between measured plasma conditions and SOLPS5.0 code calculations

“…In contrast, the measurements in reversed field give an almost stagnant plasma close to the separatrix, with an increase to M ∼ 0.2 around 1 cm and further to M ∼ 0.5 at 2.5 cm, directed towards the outer divertor. The results are consistent with earlier reports from AUG, indicating an increase in the forward field flow velocity with decreasing n e , and fall in between reports from other tokamaks, where in reversed field either large flows towards the outer target (TCV, below LFS midplane [25]) or smaller flows towards the inner target (JET, crown [26]) have been observed. The simulations agree qualitatively with the measured flow directions but, in contrast to the experiment, the modelled flows are symmetrically reversed with the field reversal.…”

“…However, also strong parallel flows, measured in the SOL of various tokamaks [24][25][26], could play a role in, for example, transporting impurities from one target to another. Despite dedicated efforts in numerical interpretation (see e.g.…”

Be wall sources and migration in L-mode discharges after Be evaporation in the JET tokamak

“…It turned out that both materials are predominantly deposited in the inner divertor [4,5]. This observation can be interpreted as a consequence of erosion dominating at main chamber wall and outer divertor strike point area and migration of eroded material to the inner divertor following the experimentally observed plasma boundary flow pattern from the low-field side to the high-field side of the boundary plasma [6][7][8]. To study the time scales of the corresponding erosion and migration processes in more detail and, like in trace impurity injection experiments, to obtain data suitable for benchmarking of impurity transport modelling, dedicated experiments with focus on a particular discharge scenario are required.…”

Particle and parallel momentum balance equations with inclusion of drifts, for modelling strong- to weakly-collisional edge plasmas

“…The motivation for improved understanding and modelling is therefore strong. Extensive theoretical and modelling effort in recent years have confirmed the significant role of drifts in both in-out asymmetries and parallel flows [22][23][24][25][26][27][28][29][30][31][32][33][34][35][36][37][38][39]. Ref.…”

“…Ref. [39] however, presents a convincing demonstration that present-day 2D edge codes are failing to predict the large parallel ion flows widely measured in the SOL. Whether this is due to improper implementation of drifts into the codes, the result of poloidal variations in turbulent transport coefficients, or caused by other, unknown reasons, is at present unclear.…”