Power and particle fluxes to plasma-facing components in mitigated-ELM H-mode discharges on JET

https://doi.org/10.1016/j.jnucmat.2010.12.015Get rights and content

Abstract

The effect of ELM-mitigation techniques such as the use of magnetic-field perturbations, gas fuelling and impurity seeding using nitrogen has been studied. With increasing ELM frequency the divertor–heat-flux factor reduces only weakly. The heat-flux factor, which has been analysed taking into account ELM-profile broadening and strike-point movement and which is proportional to the surface-temperature rise due to the heat pulse during the ELM, decreases with confinement and pedestal pressure at the same rate for mitigated ELMs using gas fuelling and magnetic-perturbation fields. Only at very high fuelling rates of nitrogen is a drastic drop of the heat-flux factor with modest confinement reduction found.

Introduction

ELMy H-mode is one of the scenarios foreseen for Q = 10 operation in ITER. The accompanying transient heat loads to the divertor in such plasmas are not only an urgent issue for ITER [1], but also for the metallic wall which is presently under construction at JET [2]. Presently, two materials are under discussion for the divertor in ITER: carbon-fibre-reinforced carbon (CFC) and tungsten. The large surface-temperature rise, ΔTsurf, at the divertor targets during ELMs can cause cracking and sublimation in the case of carbon and ablation in the case of tungsten. A heat pulse onto a solid body leads to an increase of Tsurf, which is roughly proportional to the energy deposited over the wetted area Awetted divided by the square root of the duration of the pulse tdur. In order to characterize the surface-temperature rise during an ELM we introduce the divertor–heat-flux factor ηELM:ΔTsurfηELM=EELMAwettedtdur,with EELM the energy deposited during the ELM at either the inner or outer divertor target. For the ITER-targets, cracking or melting are expected for heat-flux factors above ∼40 MJ m−2s−1/2 [3]. However, uncontrolled ELMs in ITER for the 15 MA scenario have ηELM of about 600 MJ m−2s−1/2 [4]. To reduce this potential threat for the divertor tiles various ELM-mitigation techniques are under consideration, which aim at achieving higher ELM frequencies fELM to benefit from the favorable inverse ELM-energy scaling ΔWELMfELM-1, where ΔWELM is the energy released during the ELM at the pedestal. In recent JET campaigns, resonant-magnetic-perturbation (RMP) fields [5], [6], vertical joggling (“kicks”) [7], impurity seeding [8] and gas puffing have been compared as active ELM-mitigation techniques.

Section snippets

Application of external magnetic-field perturbations (RMPs)

The JET tokamak is equipped with four external error-field-correction coils (EFCCs), which are mounted at equally-spaced toroidal locations. The EFCCs can be operated either in an n = 1 configuration, which leads to a strong core perturbation and can seed locked modes, or in an n = 2 configuration, which provides good edge ergodization. The radial perturbation field BR is for the same coil current larger at the outer plasma boundary, when the EFCCs are operated in n = 2 mode, and is of the order of 6 

Divertor target profiles

A new fast infra-red (IR) camera views the outer strike-zone in the JET divertor. Its view covers the same spatial range as Langmuir probes (LP) which are embedded into the divertor. The time resolution of the IR was 86 μs and that of the LPs 100 μs, being sufficient to resolve typical JET Type I ELMs, which have a rise time of ∼100–200 μs and a decay time of a few milliseconds. More details on the IR diagnostic are given in [9].

The target plate viewed by the IR camera is equipped with five triple

Impact of ELM energy on material limits

As pointed out earlier, the relevant quantity for material limits is the energy density and the heat-flux factor ηELM. In an earlier publication [14] the ratio of total power arriving at the divertor target over the wetted area when the power has reached its maximum has been used. However, as shown in Fig. 6, the wetted area, inferred from the total power Ptot(t) over the surface maximum heat flux Qsurf(t, R), is maximal after the ELM peak. This has to be borne in mind when calculating the

Effect of nitrogen seeding on ELM impact

Nitrogen (N2) is an attractive candidate to increase the divertor radiation, which is expected to lower the steady-state power loads to the divertor targets and the electron temperature in front of them. In high-triangularity plasmas (δave = (δlow + δup)/2 = 0.42) at a plasma current of 2.5 MA, toroidal field of 2.7T, q95 = 3.5, scans of deuterium gas fuelling (up to 2.8 × 1022 el/s) and seeding rates (up to 4.7 × 1022 el/s) have been carried out [8]. An analysis of the pedestal pressure profile revealed a

Summary and conclusions

In order to be able to compare the different ELM-mitigation techniques, gas puffing, application of external magnetic-perturbation field and seeding of extrinsic impurities, which have been applied to different target plasma, the reduction of the ELM impact factor with change in confinement has been quantified by normalising the data with respect to the corresponding reference pulse (unfuelled, no EFCC and no impurity seeding). The results are summarised in Fig. 15. Each data point of the

Acknowledgement

This work, supported by the European Communities under the contract of Association between EURATOM/Belgian State, was carried out within the framework of the European Fusion Development Agreement. The views and opinions expressed herein do not necessarily reflect those of the European Commission.

References (15)

  • N. Klimov

    J. Nucl. Mater.

    (2009)
  • G. Maddison

    J. Nucl. Mater.

    (2011)
  • T. Eich

    J. Nucl. Mater.

    (2011)
  • S. Jachmich

    J. Nucl. Mater.

    (2007)
  • G. Federici

    Plasma Phys. Control. Fusion

    (2003)
  • G.F. Matthews

    Phys. Scr.

    (2007)
  • R.A. Pitts et al., in: Proc. 19th Int. Conf. on Plasma Surface Interactions, San Diego, USA,...
There are more references available in the full text version of this article.

Cited by (21)

  • ELM mitigation via rotating resonant magnetic perturbations on MAST

    2015, Journal of Nuclear Materials
    Citation Excerpt :

    Naturally, the inclusion of additional parameters in the calculation compared to the peak heat flux increase the uncertainty in the calculated values. However, the heat flux factor (calculated using the method in [11]) is seen to reduce in the ELMs studied here, falling from 0.31 MJ m−2 s−0.5 to 0.15 MJ m−2 s−0.5 for a halving of the ELM energy, consistent with the results seen in the peak heat flux analysis. Heat flux profiles across the divertor can be used to determine the level of strike point splitting generated by the RMP.

  • Divertor load footprint of ELMs in pellet triggering and pacing experiments at JET

    2015, Journal of Nuclear Materials
    Citation Excerpt :

    In this equation, the ELM duration is defined as the interval between the sharp increase of the power deposited on the outer strike point and the typical minimum observed after the ELM (Fig.␣1). According to a recent study performed at JET [7], the divertor heat flux factor decreases very weakly with increasing ELM frequency independently of the mitigation technique employed, including resonant magnetic field perturbation (RMP), gas-fueling and impurity seeding. This, of course, reduces the mitigation potential of ELM frequency control though it can support impurity control in metallic devices such as JET in the Be/W wall configuration.

  • Divertor heat fluxes and profiles during mitigated and unmitigated edge localised modes (ELMs) on the Mega Amp Spherical Tokamak (MAST)

    2013, Journal of Nuclear Materials
    Citation Excerpt :

    The decrease in the peak heat flux during mitigation is accompanied by a decrease in the ELM wetted area, as shown in Fig. 4. The ELM wetted area is determined for the outer strike point using the definition in Jachmich et al [14]. The wetted area in natural ELMs is approximately 0.6 m2 for the largest ELMs (15 kJ), decreasing to 0.4 m2 when the ELM size decreases by a factor of three.

  • ELM mitigation techniques

    2013, Journal of Nuclear Materials
    Citation Excerpt :

    Proving this point will be very difficult without having pedestal plasmas equivalent to those in ITER. Due to the urgency of the ELM issue for ITER-class tokamaks and beyond, a rather large array of suppression and mitigation concepts have recently emerged, including techniques such as: (1) the application of edge Resonant Magnetic Perturbation (RMP) fields [4–9], (2) ELM pacing with small, high frequency, frozen deuterium pellets [10–12], (3) operations in naturally occurring quiescent H-mode (QH-mode) plasmas [13–18] and I-mode regimes [19–21], (4) impurity seeding [22–24], (5) the use of small periodic vertical equilibrium displacements [25–27], (6) the injection of supersonic molecular beams [28–30], (7) the application of low recycling wall materials like lithium coatings [31–33], (8) the application of edge Electron Cyclotron Heating (ECH) [34], (9) the introduction of small toroidal ripple fields [35,36], and (10) operations in naturally occurring small ELM regimes which include the Enhanced Dα (EDA) H-mode [37], grassy ELMs [38,39], the High Recycling Steady (HRS) H-mode [40], Type-II ELMs [41], Type V ELMs [42] and snowflake divertor configurations [43]. While several of these approaches have emerged quite recently others such as pellet pacing, operations in QH-modes, the application of edge RMP fields and a wide variety of small ELM regimes have been developed to a relatively high level of maturity over an extended period and on a broad range of tokamaks.

  • Characteristics of divertor heat and particle deposition with intrinsic and applied 3-D fields in NSTX H-mode plasmas

    2011, Journal of Nuclear Materials
    Citation Excerpt :

    However, the application of small, non-axisymmetric magnetic field perturbations produced by in-vessel or ex-vessel coils has been recently found to modify and break the axisymmetry of divertor profiles as well as to have significant impact on the Edge Localized Mode (ELM) characteristics in tokamaks [1]. As ITER presently considers the use of 3-D magnetic perturbation for the ELM suppression as a possible option [2], the effect of these 3-D fields on the heat and particle footprints on the divertor plates is of substantial interest and has recently drawn attention from various machines [3–6]. In DIII-D [7], large Type-I ELMs have been successfully eliminated [8,9] by applying resonant magnetic perturbations (RMPs) produced by a series of coils inside the vacuum vessel (internal or “I-coils”).

View all citing articles on Scopus
1

Presenting author.

2

See appendix of F. Romanelli et al., Fusion Energy 2008 (Proc. 22nd Int. Conf. Geneva) IAEA, Vienna (2008).

View full text