Nonlocal Transient Transport and Thermal Barriers in Rijnhuizen Tokamak Project Plasmas

Abstract

In the Rijnhuizen Tokamak Project plasmas, a transient rise of the core electron temperature is observed when hydrogen pellets are injected tangentially to induce fast cooling of the peripheral region. High-resolution Thomson scattering measurements show that the T e rise is associated with large temperature gradients in the region 1 , q , 2. This region acts as a layer of transiently increased thermal resistivity (transport barrier) when probed by fast heat pulses from modulated electron cyclotron heating.[S0031-9007(99)09366-7] PACS numbers: 52.55. Fa, 52.25.Fi, 52.50.Gj A widespread observation in transient transport experiments in tokamak plasmas is the so-called "nonlocal" response of the plasma core to a fast perturbation of the electron temperature ͑T e ͒ in the outer layers of tokamak plasmas [1]. Most commonly, a core T e rise is observed in response to fast edge cooling (cold pulse) by laser ablation or oblique pellet injection [1]. As reported in the literature, the central T e response contradicts local transport predictions in that (a) the core response is of opposite polarity and of larger magnitude than the edge perturbation; (b) the response takes place before the edge perturbation diffuses in. The phenomenon cannot be ascribed to Ohmic power redistribution according to [1][2][3]. Therefore it is interpreted in terms of a fast change of the core electron heat diffusivity ͑x e ͒ in response to a change in some plasma parameter elsewhere [1].A common feature is a density ͑n e ͒ dependence of the nonlocal response. At low n e , the nonlocality is strongest as the T e perturbation reverses its polarity in the plasma core. As n e is increased, the core T e variation is reduced and can be replaced by a fast response of the same polarity [1].Another common feature is a change, on the same time scale as the central T e variation, of the sawtooth period and amplitude, which has suggested a possible link between nonlocal transport and MHD behavior of the plasma [3].In this Letter we report new experimental results on the issue of nonlocal transport, obtained in the Rijnhuizen Tokamak Project (RTP). They are based on highresolution measurements of T e with Thomson scattering and on the use of modulated electron cyclotron heating (MECH) on top of cold pulses. These experiments show that the central T e rise following edge cooling takes place through the formation of a large temperature gradient (thermal barrier) in a radially localized region of the plasma. Such a barrier acts as a layer of transiently increased thermal resistivity when probed by MECH heat pulses.RTP (major radius R 0.72 m, minor radius a 0.16 m) is dedicated to transport studies and equipped with advanced diagnostics. Time resolved measurements of T e and n e are taken, respectively, with an electron cyclotron emission (ECE) radiometer covering the entire profile with 15 channels and with a 16 chord microwave interferometer. High spatial resolution is obtained with a double pulse Thomson scattering system measuring T e and n e on a vertic...

Non-local heat transport, rotation reversals and up/down impurity density asymmetries in Alcator C-Mod ohmic L-mode plasmas

“…Following the edge cooling, there was a rapid increase of the core electron temperature, on a time scale (∼5 ms) faster than the diffusive time. This effect is suggestive of a short-lived internal transport barrier (ITB), triggered by the sudden increase of the edge temperature gradient, which can be modeled with an abrupt drop in the core electron thermal conductivity [19,20,21,23,24,25,27], and which persists for the duration of the edge cooling (∼30 ms). A model for this transient increase in electron heat transport (a drop in thermal conductivity) is shown in Fig.3, with a 25% drop in χ e near r/a = 0.3 (solid to dash-dot lines).…”

Neoclassical physics in full distribution function gyrokinetics

“…These facts have been emphasised both theoretically: through the description of avalanching and spreading, [3][4][5][6][7][8][9] through the characterisation of nonlocal, nondiffusive behaviour 1,[10][11][12] and experimentally through some yet-to-be-understood experimental jigsaws: deep inconsistencies with a (fixed gradient) local and diffusive modeling have indeed been reported in perturbative (either hot or cold pulse) experiments, [13][14][15][16][17] off-axis heating experiments, 19,20 or whilst reporting Bohm-like scalings of the energy confinement time. 18 An accurate description of such dynamics requires the simultaneous and self-consistent treatment of the full gyrokinetic distribution function (full-f modeling), in full-torus (global) tokamak geometry and for a prescribed distribution of sources and sinks (flux-driven description).…”

How turbulence fronts induce plasma spin-up

“…This effect is a key for understanding why "the tail wags the plasma," i.e., why jogging edge flows by changing magnetic geometry leaves a footprint in the core flow. A perturbative experiment [23][24][25][26][27][28] is proposed as a further critical test.…”

“…Another important consequence that the nonlinear flux can induce is the dynamic response of the core flows against the edge perturbation [23][24][25][26][27][28]. As discussed above, the nonlinear flux allows the fluctuation momentum to propagate faster than the mean momentum.…”