Since 18 of December 2019 uses Nucleus credentials. Visit our help pages for information on how to Register and Sign-in using Nucleus.
May 10 – 15, 2021
Virtual Event
Europe/Vienna timezone
The Conference will be held virtually from 10-15 May 2021

H-mode physics studies on TCV supported by the EUROfusion pedestal database

May 12, 2021, 2:00 PM
4h 45m
Virtual Event

Virtual Event

Regular Poster Magnetic Fusion Experiments P4 Posters 4




Introduction – The H-mode confinement regime will be the main operational scenario on ITER and also the current foreseen scenario for fusion reactors. A continuous effort towards better predictive capabilities of H-mode confinement is being pursued on both experimental and theoretical fronts. The H-mode is characterised by the formation of a pedestal near the plasma edge and as the fusion power scales as $p_{ped}^2$, it is advantageous to maintain a high pedestal pressure. This goal is challenged by the need to mitigate H-mode characteristic Type-I ELMs either by creating a highly radiating divertor using impurity injection [1] or via ELM-free or small ELMs regimes [2]. With a pedestal database, the H-mode physics studies performed on TCV are reviewed with emphasis on the comparison between the historically ECRH dominated scenario and the NBH H-mode regime explored more recently.

Pedestal database – The TCV pedestal database is one of several databases promoted by EUROfusion to stimulate the multi-machine comparisons, (JET, AUG, MAST-U and TCV), with common parameter definitions [3] and a common platform (IMAS: ITER integrated modelling and analysis suite). The pedestal structure is determined from the pre-ELM temperature and density profiles (75-99% of the ELM cycle) using TCV’s Thomson Scattering [4]. Pedestal parameters are obtained using a mtanh fitting function [5]. To reduce uncertainties in the equilibrium reconstructions, temperature and density profiles were systematically shifted such that $T_{e,sep}=50$ eV, estimated by using the two point model for the power balance at the separatrix. To enhance the overall pedestal data quality, entries were selected according to following rules: steady state intervals over at least 0.4s (~10$\tau_E$) and a reduced $R^2$ for the fit in the region 0.8<$\psi_N$<1.05 larger than 0.75. A new plasma equilibrium is computed using the CHEASE code from the fitted pedestal, accounting for the bootstrap current and, only then, is the pedestal stability analysed [6]. The TCV pedestal database currently contains $\sim$350 entries for about 100 implemented parameters. For TCV’s heating methods, (Ohmic, ECRH and NBH), the normalised average lost energy as a function of the ELM frequency is shown Fig.1-left with the pedestal temperature plotted against the pedestal density, Fig.1-right. The plasma shaping capabilities of TCV have been exploited in the range of parameters: 1.3<$\kappa$<1.8, 0.2<$\delta_b$<0.8 and -0.3<$\delta_u$<0.8. For similar H-mode plasmas except the top triangularity ($\delta_u\sim$-0.2 vs $\delta_u\sim$0.2), the pedestal width is increased by 25% while its top pressure is reduced by a factor of 2 for the negative triangularity cases, in line with EPED predictions [6].
(left) Normalised stored energy losses vs ELM frequency; (right) Pedestal top temperature  vs pedestal top density for the 3 heating scenarios applied on TCV

ECRH ELMy H-modes – H-mode plasmas with Type-III ELMs are achieved in TCV with Ohmic heating only for $q_{95}$<3 (Fig.1 blue points). The central plasma density is too high for 2nd harmonic ECR heating, but central heating at third harmonic (X3) is possible. As the injected X3 power is increased (0<$P_{X3}$<0.5 MW), the type-III ELM frequency decreases to $f_{ELM}\sim$ 50 Hz. Additional X3 power doesn’t change the ELM frequency but increases significantly the normalised lost energy ($\Delta W/W\sim$17%). This is explained by some coupling of the ELM instability with the core MHD (m/n=1/1) that propagates the ELM crash to the plasma core region. Finally, for $P_{X3}$>0.8 MW, the ELM frequency increases, signifying a Type-I ELM regime, with a decrease in the normalised ELM losses [7-8]. Typical pedestal values for an ELMy H-mode heated with 1 MW of ECRH are $T_{e,ped}\sim$0.8 keV, $n_{e,ped }\sim3\times10^{19}$ m$^{-3}$, $n_{e,sep}\sim0.2n_{e,ped}$ and $T_e(0)/T_i(0)\sim$6. Such hot plasmas at low densities for $\psi_N$>0.8, including the pedestal, is still accessible to X2 heating and it was demonstrated that the ELM frequency can be controlled by ECRH modulation [9]. Interestingly, a steady state ELM free regime has been reached with 1.2 MW of X3 power with unfavourable $\nabla B$ configurations [10]. In today’s tokamaks, pedestal collisionalities relevant for ITER ($\nu_{\star,ped}\sim$ 0.1) might be achievable with an ECRH H-mode operational regime but conditions for partial detachment, $n_{e,sep}/n_G\sim$ 0.4 at the separatrix, are impossible.

NBH ELMy H-modes – The H-mode operational space extension towards high pedestal and larger separatrix densities has become possible with TCV’s neutral beam injector (1MW, 30keV), in operation since 2015. Moreover, ELMy H-mode can be achieved at lower plasma current ($q_{95}$>3) allowing the development of an ITER baseline scenario on TCV ($q_{95}\sim$3-3.6, $\kappa$=1.7, $\delta$=0.4) [11]. A scenario at $q_{95}\sim$4.5, $\delta$=1.5 is well established with accompanying Type-I ELMs ($f_{ELM}\sim$ 100 Hz, $\Delta W/W\sim$ 10%) and typical pedestal parameters $T_{e,ped}\sim$ 0.2 keV and $n_{e,ped}\sim4\times 10^{19}$ m$^{-3}$. The effects of $D_2$ fuelling and $N_2$ seeding on the pedestal stability and plasma confinement were investigated. Both induces an outward shift of the pedestal density relative to the pedestal temperature with a corresponding outward shift of the pedestal pressure that, in turn, reduces the peeling-ballooning stability, degrades the pedestal confinement and reduces the pedestal width [12], in line with AUG and JET results [13]. A small ELM regime with high confinement was achieved if, and only if, the separatrix plasma density was large enough ($n_{e,sep}/n_G\sim$0.3) and the magnetic configuration was close to a double null ($\delta$>0.4) [14]. The extension of the regime to $q_{95}$<4 was recently achieved. A second neutral beam (1MW, 50keV), planned by the end of 2020, will open new possibilities, not only for H-mode physics at large $\beta_N$, but also for fast particle physics.

[1] A. Kallenbach et al., “Partial detachment of high power discharges in ASDEX Upgrade” Nucl. Fusion 55 053026 (2015)
[2] J. Stober, et al., “Type II ELMy H-modes on ASDEX Upgrade with good confinement at high density” Nucl. Fusion 41 1123–34 (2001)
[4] P. Blanchard et al., “Thomson scattering measurements in the divertor region of the TCV Tokamak plasmas” JINST 14 C10038 (2019)
[5] R. Groebner et al., “Progress in quantifying the edge physics of the H mode regime in DIII-D”, Nucl. Fusion 41 1789 (2001)
[6] A. Merle, et al., “Pedestal properties of H-modes with negative triangularity using the EPED-CH model”, Plasma Phys. Control. Fusion, 59, 10, 104001, 2017.
[7] A. Pitzschke, “Pedestal Characteristics and MHD Stability of H-Mode Plasmas in TCV”, PhD thesis, EPFL, no 4917 (2011)
[8] A. Pitzschke et al., “Electron temperature and density profile evolution during the ELM cycle in ohmic and EC-heated H-mode plasmas in TCV”, Plasma Phys. Control. Fusion, 54, 015007 (2012)
[9] J.X. Rossel et al., “Edge-localized mode control by electron cyclotron waves in a tokamak plasma” Nucl. Fusion 52 032004 (2012)
[10] L. Porte et al., “Plasma dynamics with second and third-harmonic ECRH and access to quasi-stationary ELM-free H-mode on TCV” Nucl. Fusion, 47, 8, 952–960 (2007)
[11] O. Sauter et al., “ITER baseline scenario investigations on TCV and comparison with AUG”, this conference
[12] U. A. Sheikh et al., “Pedestal structure and energy confinement studies on TCV” Plasma Phys. Control. Fusion, 61, 1, 014002 (2019)
[13] L. Frassinetti et al, “Role of the pedestal position on the pedestal performance in AUG, JET-ILW and TCV and implications for ITER” Nucl. Fusion, 59, 7 076038 (2019)
[14] B. Labit et al., “Dependence on plasma shape and plasma fuelling for small edge-localized mode regimes in TCV and ASDEX Upgrade” Nucl. Fusion, 59, 8 086020 (2019)

Affiliation Ecole Polytechnique Fédérale de Lausanne
Country or International Organization Switzerland

Primary author



Stefano Coda (CRPP-EPFL) Basil Duval (Ecole Polytechnique Fédérale de Lausanne – Swiss Plasma Center (SPC)) Antoine Merle (Ecole Polytechnique Fédérale de Lausanne (EPFL), Centre de Recherches en Physique des Plasmas (CRPP), CH-1015 Lausanne, Switzerland) Laurie Porte (CRPP-EPFL) Olivier Sauter Umar Sheikh Mike Dunne (IPP-Garching) Lorenzo Frassinetti (KTH, Royal Institute of Technology) Rory Scannell (Association CCFE/Euratom)

Presentation materials