Impact of the pedestal plasma density on dynamics of edge localized mode crashes and energy loss scaling

Abstract

The latest BOUTþþ studies show an emerging understanding of dynamics of edge localized mode (ELM) crashes and the consistent collisionality scaling of ELM energy losses with the world multitokamak database. A series of BOUTþþ simulations are conducted to investigate the scaling characteristics of the ELM energy losses vs collisionality via a density scan. Linear results demonstrate that as the pedestal collisionality decreases, the growth rate of the peeling-ballooning modes decreases for high n but increases for low n (1 < n < 5), therefore the width of the growth rate spectrum c(n) becomes narrower and the peak growth shifts to lower n. Nonlinear BOUTþþ simulations show a two-stage process of ELM crash evolution of (i) initial bursts of pressure blob and void creation and (ii) inward void propagation. The inward void propagation stirs the top of pedestal plasma and yields an increasing ELM size with decreasing collisionality after a series of microbursts. The pedestal plasma density plays a major role in determining the ELM energy loss through its effect on the edge bootstrap current and ion diamagnetic stabilization. The critical trend emerges as a transition (1) linearly from ballooning-dominated states at high collisionality to peelingdominated states at low collisionality with decreasing density and (2) nonlinearly from turbulence spreading dynamics at high collisionality into avalanche-like dynamics at low collisionality. V C 2014 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4905070]The phenomena of nonlocal transport, such as avalanches or turbulence spreading, are well-known in diverse systems. Examples include, but are not limited to: piles of granular matter, discrete energy dissipating events in earthquakes and solar flares, and turbulent overshoot and penetration in fluid turbulence. Nonlocal dynamics of turbulence, transport, and zonal flows are often observed in plasma turbulence simulations and experiments, such as Edge Localized Modes (ELMs) in tokamaks. ELMs are a common characteristic feature of the tokamak H-mode plasma regime, where the high frequency ELM instability repeats periodically throughout the high confinement mode (H-mode) phase of the discharge. 1 The instability causes quasi-periodic relaxations of the edge pedestal, resulting in a series of hot plasma eruptions on a fast MHD timescale and leading to large energy fluxes to the plasma facing components (PFCs), which will suffer from excessive ablation, fast erosion, or melting. The concern about the survival of PFCs in ITER has sparked intense interest in ELM dynamics and in the parameter scaling of ELM energy loss. Numerous experiments in divertor tokamaks have shown a decrease in the relative type I ELM energy loss with increasing pedestal density (collisionality) over a decade ago. 2 There are attempts to provide an explanation for the dependence of ELM energy loss on collisionality. 3 However, there is as yet no common accepted explanation of the observed scaling neither from analytical theory nor numerical simulations. 4He...

The impact of collisionality, FLR, and parallel closure effects on instabilities in the tokamak pedestal: Numerical studies with the NIMROD code

“…This contrasts with Ref. [46] that examines these effects in concert and Ref. [47] that considers only the effect of modifications to the current profile.…”

Impact of inward turbulence spreading on energy loss of edge-localized modes

“…So, the shape of the growth rate spectrum cðnÞ and the toroidal mode number n corresponding to the maximum growth rates are almost the same during the b scan, which is different from the density scan for a fixed pressure where both density and temperature change. 13 Because g i ¼ 0:685, which is too small to drive the KBM instability below the ideal ballooning unstable threshold, the unstable thresholds of IBM and KBM are similar. The red dashed line shows the effect of drift Alfv en wave (DAW).…”

“…The theory and the numerical implementation in codes such as ELITE 3,4 show that the type-I ELM is triggered by ideal peeling-ballooning (P-B) modes, which are driven by the large pressure gradient and current profile in the pedestal region. The BOUTþþ code [5][6][7][8][9][10][11][12][13] has successfully simulated linear growth and nonlinear crash phase of ELMs in circular and advanced divertor geometries. The P-B turbulence before an ELM crash has been proven to be important for the trigger of ELMs.…”

E × B flow shear mitigates ballooning-driven edge-localized modes at high collisionality: experiment and simulation

“…The bicoherence of the coherent mode in the co-NBI case (f = 32kHz) and ctr-NBI case (f = 42kHz) are illustrated in the Figure 4 been described in detail in Ref. [15]. The non-ideal physics effects include diamagnetic drift, E × B drift for typical pedestal plasmas.…”

“…The normalized collisionality on the top of the pedestal is ν * = 2.3.Here, n e,ped = 3.5 × 10 19 m −3 and T e,ped = 300eV . Comparing to the previous set-up in BOUT++ simulations in Ref [15],. the collisionality on EAST is much higher than the n 0 = 5 ∼ 9 × 10 19 m −3 cases with ν * = 0.2 ∼ 0.8, but close to the n 0 = 20 × 10 19 m −3 case with ν * 6.19.…”