### Speaker

### Description

$\qquad$ Characterizing and understanding the power threshold conditions for ITER to achieve H-modes ($P_{LH}$) is a major goal of a series of L-H transition experiments undertaken at JET since the installation of the ITER-like-wall (JET-ILW), with Beryllium wall tiles and Tungsten divertor [1,2,3,4]. In this contribution we report on results from L-H transitions studies in H, D and new almost pure $^4Helium$ plasmas, and compare the results with ITER predictions. The most notable result is that the density at which $P_{LH}$ is minimum, $n_{e,min}$, is considerably higher for $^4He$ than for D, and strongly influenced by shape.

$\qquad$ A detailed analysis of the pre-transition $E_r$ profiles across the ne scan in D and $^4He$ find matching qualitative changes in the $E_r$ profile. In high field NBI heated D plasmas, we report on power balance analysis and its impact on $n_{e,min}$. Modelling of the plasma SOL does show differences in the heat flux required to drive a transition between H and D (in the high $n_e$ branch), and $^4He$ plasmas are also being studied.

**Characterizing the L-H transition power threshold for H, D, $^4He$: $n_{e,min}$, ion heat flux, $E_r$**

$\qquad$ The interest on $^4He$ plasmas is not purely academic, and our data brings surprises. The ITER Research Plan includes a low toroidal field Pre-Fusion Operating Power phase with either Hydrogen or Helium plasmas in order to study H-modes as early as possible, before the nuclear phase that starts with D plasmas. A prediction of $n_{e,min}$, was made inspired on the studies of Ryter [5], who observed in AUG that a sufficient edge ion heat flux is necessary to achieve a sufficient radial electric field (shear). Assuming pure electron heating in ITER, $n_{e,min}$ has been evaluated on the basis of 1.5-D transport modelling as the density at which the ratio of edge ion power flux to total edge power flux starts to saturate with increasing density. The result of this modelling is that $n_{e,min}$~$0.4 n_{GW}$, independent of the ion species [6]. The transition condition in that model is itself based on the assumption that the He power threshold, $P_{LH}(He)$, is 1.4 x $P_{LH}(D)$, while $P_{LH}(H)$=2x$P_{LH}(D)$ [7].

$\qquad$ We find that for Horizontal Target plasmas the estimate of $n_{e,min} \sim 0.4 \times n_{GW}$ is in agreement with the D data, but $n_{e,min}$ is closer to $0.5 \times n_{GW}$ for H, and to 0.6 for Helium. NBI heated H plasmas have higher $P_{LH}$ than RF heated ones, the reasons are still being investigated. The data points with no black outline correspond to transitions during an NBI blip, so $P_{LH}$ is probably higher. In fact radiation is considerably higher for the dominantly RF-heated Horizontal Target Helium plasmas at low density, so the auxiliary power required for the L-H transition to take place is lower for H than for He below 3.3×$10^{19} e/m^3$. Above $n_{e,min}$(He), D and He have similar $P_{LH}$, below the Martin scaling, while H has a much higher $P_{LH}$.

In the 1.8 T Corner dataset $n_{e,min}$ is not so easily identified. Above 0.4$n_{GW}$, $P_{LH}$ in Corner is generally higher than in Horizontal Target for all species, approaching the Martin scaling for D and He, much higher for H.

$\qquad$ The strong shape effect shown in all L-H transition datasets at JET is in apparent contradiction with the ion heat channel determining $n_{e,min}$. A detailed study of the relation between ion heat flux and $n_{e,min}$ in 3T, 2.5 MA D plasmas, now with $T_i$ measurements, is underway [9]. We find the e-i exchange term is subdominant and unlikely to determine $n_{e,min}$.

$\qquad$ In a dataset with Horizontal Target, 2.4 T, 2 MA, NBI-heated plasmas (not shown), we find that $n_{e,min}(He) \sim 0.7 \times n_{GW}$, while $n_{e,min}(D) \sim 0.4 \times n_{GW}$. Above $n_{e,min}(He)$, D and He have similar $P_{LH}$. In this case we are attempting to reproduce the ITER transport models and $P_{LH}$ predictions and contrast them with the data. For these plasmas Doppler reflectometry shows that the $E_r$ profile of the low ne branch for both D and He plasmas has a modest $E_r$ well inboard of the separatrix and a sharp peak further out, while the high density branch has a clear $E_r$ well, but no peak near the separatrix.

$\qquad$ DIII-D results show a ~30% increase in $n_{e,min}$ of He plasmas relative to D [10], lower than our 50% shift. AUG studies show no difference in $n_{e,min}$ between H, D and He, and the same $P_{LH}$ for D and He plasmas[11]. In AUG $H+^4He$ mixtures [14], more than 20% $n_{He}/(n_{He}+n_D)$ is needed to see a change in $P_{LH}$(H), while <10% suffices in JET NBI heated plasmas [4]. C-Mod results [12] show He data in the low $n_e$ branch for $n_e<0.3 \times n_{GW}$, while in D $P_{LH}$ increases with density, indicating a shift in $n_{e,min}$. Above $n_{e,min}(He)$, $P_{LH}$ in JET-ILW is similar for D and He, therefore the increase in $P_{LH}$ due to higher $n_{e,min}$ is compensated by the lower power required to access it, since ITER had assumed $P_{LH}(He)$=$1.4\times P_{LH}$(D).

Simulations of L-H transitions for hydrogen isotopes with the HESEL[15] model find that $P_{LH}$ decreases with increasing mass number like $A^{-1.2}$. Results in $^4{He}$ plasmas are expected soon.

**Summary and Outlook:**

Our results question the logic that supports He for access to H-mode in the early operating phase of ITER, but not necessarily the final power estimate. Detailed analysis is on-going, to provide better understanding of the mechanisms involved and produce an improved prediction. Novel $E_r$ measurements will enable a more detailed understanding of L-H transitions in D and $^4He$ plasmas.

A Tritium campaign is planned at JET for summer 2020. We expect to obtain L-H transition data for pure Tritium, H+T and H+$^4He$ mixtures. This should inform future experiments in JET and ITER.

**References:**

[1] CF Maggi et al 2014 Nucl. Fusion 54 023007

[2] E Delabie, Proc. of the 25th IAEA FEC, Saint Petersburg, Russia, EX/P5 (2014)

[3] J Hillesheim et al Proc. of the 26th IAEA FEC, Kyoto, Japan, (2016)

[4] J Hillesheim et al Proc. of the 27th IAEA FEC, Gandhinagar, India (2018)

[5] F Ryter et al 2013 Nucl. Fusion 53 113003

[6] ITER Research Plan within Staged Approach, ITR-Report 18-003 (2018) p 351

[7] D McDonald et al, Plasma Phys. Control. Fusion 46 519 (2004)

[8] U. Kruezi et al 2020 JINST 15 C01032

[9] P. Vincenzi et al., 46th EPS conf., P2.1081 , Milano, Italy (2019);

[10] Gohil et al, Nucl. Fusion 51 (2011) 10;

[11] F Ryter et al, Nucl. Fusion 54 083003 (2014);

[12] C.E. Kessel et al 2018 Nucl. Fusion 58 056007;

[13] C. Silva, to be presented at EPS 2020;

[14] U Planck, submitted to NF

[15] Nielsen AH et al, Physics Letters A 379 (2015) 3097–3101.

Affiliation | Laboratorio Nacional de Fusion, CIEMAT |
---|---|

Country or International Organization | Spain |