The quasi-continuous exhaust regime in JET

LETTER • OPEN ACCESS The quasi-continuous exhaust regime in JET To cite this article: M. Faitsch et al 2025 Nucl. Fusion 65 024003 View the article online for updates and enhancements. You may also like Multi-functional code for hydrogen isotopes transport analyses: verification & validation against fusion-relevant applications F. Hattab, V. Narcisi, C. Ciurluini et al. - Hybrid hydrogen-electricity production using spherical tokamaks: a cost-driver sensitivity study and techno-economic analysis J. Hidalgo-Salaverri, T. Griffiths, Z. Xuereb Conti et al. - Impact of trapping on tritium selfsufficiency and tritium inventories in fusion power plant fuel cycles Samuele Meschini, Rémi DelaporteMathurin, George R. Tynan et al. - This content was downloaded from IP address 193.147.173.203 on 28/04/2025 at 10:47 International Atomic Energy Agency Nuclear Fusion Nucl. Fusion 65 (2025) 024003 (9pp) https://doi.org/10.1088/1741-4326/adaa86 Letter The quasi-continuous exhaust regime in JET M. Faitsch1,∗, M. Dunne1, E. Lerche2,3, P. Lomas3, I. Balboa3, P. Bilkova4, P. Bohm4, A. Kappatou1, D. Kos3, B. Labit5, S. Menmuir3, O. Sauter5, S. Silburn3, E.R. Solano7, H.J. Sun3, A. Tookey3, E. Viezzer6, U. Stroth1,8and JET Contributorsaand the EUROfusion Tokamak Exploitation Teamb 1Max-Planck-Institute for Plasma Physics, Boltzmannstr. 2, D-85748 Garching, Germany 2Laboratory for Plasma Physics LPP-ERM/KMS, B-1000 Brussels, Belgium 3United Kingdom Atomic Energy Authority, Culham Science Centre, Abingdon, Oxon OX14 3DB, United Kingdom of Great Britain and Northern Ireland 4Institute of Plasma Physics of the CAS, Prague, Czech Republic 5Ecole Polytechnique Federale de Lausanne, Swiss Plasma Center, CH-1015 Lausanne, Switzerland 6Department of Atomic, Molecular and Nuclear Physics, University of Seville, Seville, Spain 7Laboratorio Nacional de Fusión, CIEMAT, Madrid, Spain 8Physik Department E28, Technische Universität München, 85748 Garching, Germany E-mail: [email protected] Received 25 September 2024, revised 12 December 2024 Accepted for publication 15 January 2025 Published 22 January 2025 Abstract The quasi-continuous exhaust (QCE) regime is a regime that is naturally type-I ELM-free. It combines the high density at the plasma edge needed for power exhaust with the high normalised energy confinement typical for H-mode operation. In the QCE regime large-scale ELMs are avoided and high-frequency, low-amplitude filaments are present leading to the name-giving quasi-continuous edge transport of particles and energy. This contribution reports that for the first time the QCE regime was successfully achieved in JET with a metal wall. Moreover, it was demonstrated in the recent JET deuterium-tritium campaign DTE3 that the regime is compatible with D–T operation. Porting the QCE regime to JET strongly benefited from the experimental and modelling efforts at the medium sized tokamaks ASDEX Upgrade and TCV. Using the physics picture developed from the ASDEX Upgrade experimental results, the route to the QCE regime in JET reported here is following closely the approach that was successful in ASDEX Upgrade. First, strong plasma shaping—large elongation and triangularity and the highly correlated closeness to double null—is developed. Second, sufficient fuelling to achieve high enough density at the pedestal foot, close to the separatrix, is applied. In addition, neon seeding proved to be very beneficial to avoid type-I ELMs when reducing the main ion fuelling. aSee Maggi et al 2024 (https://doi.org/10.1088/1741-4326/ad3e16) for JET Contributors. bSee Joffrin et al 2024 (https://doi.org/10.1088/1741-4326/ad2be4) for the EUROfusion Tokamak Exploitation Team. ∗Author to whom any correspondence should be addressed. Original Content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. 1741-4326/25/024003+9$33.00 Printed in the UK 1 © 2025 The Author(s). Published by IOP Publishing Ltd on behalf of the IAEA Nucl. Fusion 65 (2025) 024003 M. Faitsch et al Keywords: power exhaust, ELM-free, quasi-continuous exhaust, deuterium-tritium, plasma edge (Some figures may appear in colour only in the online journal) 1. Introduction An integrated reactor scenario combines good core conditions to achieve high fusion gain together with a plasma edge that guarantees a safe power exhaust solution. For many decades, the high confinement mode with type-I edge localised modes (type-I ELMy H-mode) was the preferred scenario [1]. However, it became evident that the transient first wall loads due to type-I ELMs will be unacceptable for a reactor [2,3]. The active suppression of ELMs using magnetic perturbation coils was investigated with success as a possible solution [4]. However, as divertor detachment seems difficult to achieve with this method, other solutions have recently been looked for and the investigation of regimes that are naturally typeI ELM-free came into focus. An overview of various type-I ELM-free regimes is presented in [5,6]. The quasi-continuous exhaust (QCE) regime is one candidate for an integrated reactor scenario without large transients [7]. The regime is also known as the type-II ELM [8–13] or a small ELM [14,15] regime and is closely related to the Enhanced D-Alpha (EDA) H-mode [16]. The QCE regime is characterised by modes at the pedestal foot that produce highfrequency and low-amplitude filaments in the scrape-off layer [7,17]. The access conditions are strong plasma shaping and high fuelling. In addition to the absence of large-scale transients, the confinement time and pressure are similar to those of the type-I ELMy H-mode [18]. Significant progress in the understanding of the QCE regime has been achieved in recent years. Simulations with the HELENA stability code indicate that local ballooning modes become unstable at the pedestal foot. They induce a local flattening of the pressure gradient that stabilises the global peeling-ballooning modes causing type-I ELMs [19–21]. A similar local flattening follows from the turbulence parameter αt, defined in [22], which is proportional to the pedestal foot collisionality [18]. This parameter is closely linked to resistive ballooning modes [23]. Due to strong gas fuelling, the pedestal density, collisionality and thus αtare relatively high, also hinting at the importance of resistive turbulence at the edge of QCE pulses. The small instability characteristic for the QCE regime is equivalent to type-II ELMs which were already observed in JET with the carbon wall [9,10]. However, these previous experiments showed only transient phases without type-I ELMs while stationary phases are routinely obtained in AUG and TCV. With the newly gained insight from AUG [7,14, 18–21] and TCV [15,24], a new attempt at realising stationary, type-I ELM-free QCE plasmas was started in JET now equipped with tungsten divertor and beryllium first wall [25]. The key objectives of these experiments were (i) to realise a stationary QCE regime, (ii) to decrease the pedestal top collisionality compared to the experiments in AUG and TCV where ν∗ e,ped is of the order of 1–10 while in ITER values in the order of 0.1 are expected, and (iii) to prove that the regime is compatible with deuterium–tritium (D–T) plasma operation. Achieving low pedestal top collisionality is a long-term outstanding challenge of QCE research. As the QCE regime is naturally operated at high normalised density, only a large device with high heating power can simultaneously achieve low pedestal top collisionality. The paper is organised as follows: section 2introduces the key experimental conditions to reach stationary QCE phases with appropriate shaping, fuelling and neon seeding. In section 3the compatibility of the QCE regime with a deuterium-tritium plasma is presented. A discussion on the achieved parameter ranges and conclusions follow in sections 4and 5, respectively. 2. Stationary QCE regime 2.1. Experimental approach In JET the QCE regime is achieved at high plasma shaping as in AUG and TCV with high elongation, high triangularity and, linked to this, closeness to double null as shown in figure 1. A significant effort for the shape development was invested. The shape is based on previous attempts to achieve a typeII ELM scenario in JET with carbon wall [9]. The discharge with the shape in figure 1, JPN 102902, has Ip=1.5 MA and |Btor|=2.3 T. The magnetic equilibrium is reconstructed using the EFIT code with pressure constraints [26]. Shown in red is the first wall structure, in black the separatrix and in blue the flux surfaces inside the confined region as well as in the scrapeoff layer. The secondary X-point is close to the top of the first wall structure. The shaping parameters are κ=1.83,δlow =0.44,δup =0.47,dR,XP ≈15mm with elongation κ, upper and lower triangularity δand distance between the primary and secondary separatrix at the mid-plane dR,XP. These values are close to the projected values of ITER discharges with κ=1.85 and δ=0.49 [27]. Despite the proximity to double null, no significant heat flux to the upper part of the vessel was observed in the parameter range studied, however, visible light is emitted showing that some recycling flux reaches the top of the tokamak. The active lower divertor is in a vertical inner and horizontal outer target configuration chosen to maximise diagnostics coverage and accommodate the high plasma shaping; the energy limit of the tungsten outer divertor is the main limiting factor for the discharge duration [28]. Figure 2shows the typical time traces of the experiments. Presented are an ELMy, JPN 103404, and a QCE, JPN 103451, discharge with identical engineering parameters except for the fuelling rate. More details on the fuelling dependence are 2 Nucl. Fusion 65 (2025) 024003 M. Faitsch et al Figure 1. Poloidal cross-section illustrating the close-to-double-null shape used in the QCE experiments. The secondary X-point is close to the upper vessel structure with dR,XP ≈15mm. Shown in magenta is the line of sight of the visible spectroscopy used for the Be-II and W-I line emission. given in section 2.3. The heating power is increased step-wise and the H-mode is entered with the second step at t≈11.5 s. The QCE-compatible plasma shape (shown in figure 1) is reached around t≈13.0 s. About 0.5 s later, the originally large type-I ELM footprint in the divertor (measured by infrared cameras) is virtually suppressed in the QCE discharge (JPN 103451) and the peak divertor temperature stays approximately constant, indicating a strong reduction in the intraELM heat-loads on the divertor tiles. Conversely, in the ELMy discharge (JPN 103404) the type-I ELMs are visible in the IR measurements throughout the discharge and the peak divertor temperature increases gradually until the end of the Hmode phase. The discharges are terminated by a slow stepwise ramp-down of the heating power and shaping from 16.4 s allowing the density to drop before exiting the H-mode phase. Note that the ELM footprint briefly re-appears in the divertor IR measurements in JPN 103451 once the transition out of the QCE shape is completed (t≈16.8 s). Both discharges have a moderate confinement scaling factor of H98y2 ≈0.75. The absence of type-I ELMs is monitored by multiple diagnostics that are sensitive to the rapid expulsion of particles and heat from the plasma edge that eventually reach the firstwall and divertor structures. Figure 3(a) shows a time window early in the pulse after the H-mode entry (around 11.5 s) using Figure 2. Time traces of global plasma parameters for an ELMy reference pulse in red and a QCE pulse in black. (A) plasma current, (B) NBI (solid) and ICRF (dashed) heating power, (C) Greenwald density fraction of the line averaged density, (D) energy confinement factor (E) shaping parameter as defined in (1) and (F) peak surface temperature of the outer divertor target. IR data is available only after 12 s for the pulse in black. a conventional strong shaping for JET (δup =0.40, κ=1.75, dR,XP ≈20mm). The typical ELM monitors, Be-II and WI line emission in the outer divertor, are shown in (A) and (B). The ELM frequency is fELM ≈110 Hz. The ELMs have a noticeable impact on the pedestal top electron temperature as measured by ECE (using a fixed channel close to ρpol =0.9) in (C) and in the outer divertor target plate temperature as measured by IR thermography [29] in (D). Figure 3(b) shows a time window during the final phase with close-to-double-null shaping (see figure 1). The Be-II signal shows clear excursions but the pedestal top electron temperature and the target temperature are virtually insensitive to the plasma edge relaxations. We call these excursions on the Be-II emission filaments; they occur with a frequency of ffil ≈300 Hz in this time window. The filaments are still visible 3 Nucl. Fusion 65 (2025) 024003 M. Faitsch et al Figure 3. Time traces of typical ELM indicators. The pulse has Ip=1.5MA, Btor =2.3T, Pheat ≈19MW and gas rate ΓD el ≈3.0×1022 s−1. (a) ELMy with conventional shape and (b) QCE with new shape. The sub-plots (A) and (B) are line emissions for beryllium and tungsten along a line-of-sight into the outer divertor region. (C) is the electron temperature for an ECE channel close to the pedestal top (ρpol =0.9) and (D) is the peak surface temperature of the outer divertor target measured with IR subtracting the minimum temperature as indicated above the panel. A running mean filter of 1.2 ms is applied to all signals to reduce the noise level. in the W–I line emission as well as in the Langmuir probe ion saturation current (not shown). We conclude that these filaments expel plasma from the confined region without affecting the pedestal top values and with only a small amount of energy reaching the divertor. However, they still cause a measurable ion flux to the divertor that may increase the sputtering of tungsten in the conditions presented here with a high-recycling but attached divertor. Future work needs to address the power load and sputtering related to the filaments in larger tokamaks. Furthermore, the compatibility of the QCE regime with detachment remains to be demonstrated also during the filaments. 2.2. Shape dependence Strong shaping is crucial to achieve the QCE regime. Local ballooning modes become unstable above a critical pressure gradient. This critical value is given by Dunne et al [21]: αcrit =0.64κ2.2(1+δ)0.9(1) Sd=κ2.2(1+δ)0.9,(2) similar to the critical value proposed by Bernard et al [30]. A proposed window for ELM suppression opens with Sd>3–4 [21]. To demonstrate that a sufficient shaping is necessary, a slow shaping ramp was performed at otherwise constant parameters. Figure 4shows time traces of JPN 103924, with the shaping variation in (A) and the outer divertor temperature in (B). The gas fuelling rate is constant at Γel ≈4.8×1022 s−1. When the shaping value decreases below the critical value of Sd≈5.2, type-I ELMs re-appear as visible in (B). Note that there occurs already a single large type-I ELM just before 15 s showing that the fuelling level is already marginal at slightly higher shaping. Figure 4. Time traces for a shaping ramp from QCE to ELMy. The pulse has Ip=2.0MA, Btor =2.8T, Pheat ≈22MW and Γel ≈4.8×1022 s−1. The transition occurs at about 15.5 s as indicated by the vertical black line. (A) shaping parameter as defined in (1) from [21] and (B) peak surface temperature of the outer divertor target. 2.3. Fuelling dependence Whilst excessive fuelling might lead to a reduced performance [31] or to an H-L back-transition (and potentially to a subsequent disruption) [32,33], high fuelling is beneficial for power exhaust as it increases the separatrix density and with this the scrape-off layer radiation [34,35]. In addition, a lower limit of fuelling to avoid type-I ELMs and staying in QCE is reported in AUG [18]. As mentioned in the introduction, the fuelling dependence can be viewed as a critical density at the very edge of the plasma. Due to the relatively fixed separatrix temperature set by parallel electron conduction, the density is the main parameter influencing both pressure and collisionality. Preliminary analysis indicates that Te,sep ≈110– 130 eV while ne,sep ≈1.5–3.5×1019 m−3in the JET data set, the detailed analysis of the separatrix parameters is still ongoing. We note that a variation in temperature has a strong impact for the collisionality as shown by [22] and in [18] for QCE in 4 Nucl. Fusion 65 (2025) 024003 M. Faitsch et al Figure 5. Time traces of (A) Be-II line emission and (B) peak outer divertor target temperature as typical ELM indicators. All pulses have Ip=2.0MA, Btor =2.8T and Sd≈5.3. AUG. We focus here on the influence of fuelling as external control parameter and show that indeed also in JET the behaviour of the QCE regime is qualitatively consistent with that in AUG and TCV. Figure 5shows the two primary ELM monitors, Be-II line emission and target temperature, for four plasma pulses with Ip=2.0 MA and |Btor|=2.8 T. The main difference between the two pulses in (a) and (b) is the fuelling rate, with ΓD el ≈3.6×1022 s−1in pulse JPN 103404 (a) exhibiting ELM behaviour and ΓD el ≈4.9×1022 s−1in JPN 103451 (b) showing the filamentary signature typical for the QCE regime. More fuelling was needed at higher plasma current and/or lower safety factor in order to obtain QCE. This is expected assuming local ballooning modes are responsible for the absence of the type-I ELMs due to the dependence on the safety factor and expected steeper gradients at higher current [18,21]. 2.4. Neon seeding Neon is a primary candidate as a radiating species in ITER [36]. In the JET experiments reported here, small to moderate amounts of Ne were used to prove the compatibility with radiating plasma conditions and to reduce the divertor heat flux. Figure 5(c) shows ELM and filament signals for a pulse with a small amount of neon injection (JPN 103407) with otherwise the same parameters as pulse JPN 103404, shown in figure 5(a). Both pulses have a deuterium injection of ΓD el ≈ 3.6×1022 s−1while in the neon seeded case a neon influx of ΓNe el ≈0.12 ×1022 s−1is added, leading to a core neon concentration of cNe ≈0.8%. Note that Ne fluxes for detachment studies are typically about one order of magnitude higher leading to concentrations inside the pedestal of the order of 2.0% [37,38]. Neon is radiating primarily in the divertor and plasma edge regions [37,38] with a peak in the cooling factor around 30 eV [39]. The ELM/filament behaviour as well as the kinetic profiles are very similar between the neon seeded discharge and those with a higher deuterium gas rate as shown in figure 5(b). The amount of additional electrons introduced by Ne is too small to change the edge density sufficiently. We speculate, that the increased radiation might sufficiently lower the power flowing through the pedestal into the scrape-off layer and thus reduce the temperature and increase the collisionality at the pedestal foot. In addition, the colder divertor might increase the efficiency of fuelling due to recycled deuterium. Both ideas are supported by IR measurements, which indicate reduced divertor target power loads. Operating at high gas rate is hampered by limits of the neutral pressure in the NBI duct in JET. Thus, substituting deuterium with neon in order to achieve QCE has a beneficial consequence for the operation of the NBI system. The lower neutral pressure within the NBI duct as well as the lower coupling resistance of the ICRF antennas indicate that the far-SOL density of the neon seeded case is lower than with the increased deuterium rate. While the ion saturation current measured at the outer divertor target does not show a clear change, the radiated power increases significantly from about Prad ≈5.8 MW to Prad ≈10.5 MW with the addition of neon at the same deuterium gas rate. The divertor target heat flux reduces 5 Nucl. Fusion 65 (2025) 024003 M. Faitsch et al accordingly. This is inferred by IR thermography showing an increase of the peak divertor temperature of ∆T≈130Ks−1 without neon and ∆T≈40Ks−1with neon averaged over a 1.5 s long flat-top phase with constant strike line position. From this we conclude that adding neon leads to conditions close to power detachment, without obtaining a reduction in the particle flux. 3. QCE operation with a deuterium-tritium mixture JET offers the unique opportunity to operate in tritium (T) and in a mixture of deuterium and tritium (D–T). Two D– T experimental campaigns were recently conducted, named DTE2 and DTE3. While DTE2 in 2021 focused among other fusion issues on the optimisation of fusion power and energy [40], DTE3 in 2023 was more focused on integrated scenarios and power exhaust investigations [41]. Studies in these campaigns showed that the near scrape-off layer power falloff length does not change outside of measurement uncertainties between D and T [42] while the density may decay more slowly, leading to an increased density in the far-SOL in T and D–T [43]. An increased plasma pressure was routinely observed in D–T and T compared to D. The main reason is an increased pedestal top pressure due to an increase in the density at roughly constant temperature, see e.g. [44,45] for type-I ELMy H-mode. The QCE experiments were one of the executed type-I ELM-free scenarios in DTE3 [41]. The main objective was to demonstrate the QCE regime in a D–T plasma and identify potential differences with D plasmas in terms of required gas puff to avoid type-I ELMs, pedestal profiles and overall confinement. As discussed, the main QCE access conditions are plasma shaping and the profiles at the very edge of the plasma. We fix here the shaping (Sd≈5.3, see cross-section in figure 1) and use a close engineering match between D and D–T plasma pulses to show that we can transfer a D plasma pulse into a D– T plasma pulse without losing the QCE access. Time traces for the D reference pulse, JPN 103446, and a pulse with an almost even mixture of D–T (with the T ratio increasing over the pulse from about 40% to 55%), JPN 104494, are shown in figure 6. They are operated with Ip=2.0 MA (A) and |Btor|=2.8 T (not shown). The heating power differs moderately due to the available heating power per individual NBI source, with PDD heat ≈23MW and PDT heat ≈ 25MW with D-only NBI, see figure 6(B). Both pulses have neon seeding leading to a very similar level and distribution of radiation, the total radiated power is shown in figure 6(C), while the core neon concentration is lower in D–T with cDT Ne ≈ 0.6%compared to cDD Ne ≈1.0%. An increased normalised plasma pressure is observed in D– T compared to D, figure 6(D) as was routinely observed over a wide range of H-mode experiments. While for most scenarios the increase in pressure was nearly exclusively attributed to an Figure 6. Time traces of global plasma parameters for a D reference pulse in blue and a D–T pulse in gold. (A) plasma current, (B) NBI (solid) and ICRF (dashed) heating power, (C) radiated power, (D) normalised plasma pressure and (E) shaping parameter as defined in (1). increase of the pedestal top density, we observe in QCE that particularly also the pedestal top electron and ion temperatures are elevated. Figure 7shows a comparison between electron and ion temperature, electron density and electron pressure profiles in the pedestal region for the two pulses during the flat-top. Apart from the higher pressure at similar heating power levels in D–T compared to D, the behaviour was very similar, e.g. the same shaping and gas amount led to the same ELM/QCE behaviour. Figure 5(d) shows the ELM signals for the D–T pulse JPN 104494. The filament characteristics are very similar to the D pulse JPN 103446 (see comparable pulse JPN 103407 in figure 5(c)). Indeed, no significant change in behaviour was expected as the main theoretical considerations do not have any [21] or only a weak [18] mass dependence. It has to be noted that due to the limited amount of experiments in D–T, small changes that are outside the probed experimental conditions might exist. Finally, one should be aware that the main purpose of the D–T fuel mix is to achieve fusion. The QCE experiments were not conducted to study the fusion processes, since the QCE regime is a high density regime, which has as a consequence moderate central temperature in JET and with this a low fusion cross-section. The main focus here was to demonstrate that a D–T plasma behaves similarly to a pure D plasma in terms of access to QCE and absence of type-I ELMs. 6 Nucl. Fusion 65 (2025) 024003 M. Faitsch et al Figure 7. Pedestal kinetic profiles for a D pulse in blue and a D–T pulse in gold, (A) electron (dots and fit) and ion (diamonds with black outline) temperature, (B) electron density and (C) electron pressure. Table 1. Parameter range of steady-state (at least 400 ms) QCE phases. min max Ip(MA) 1.5 2.25 Btor (T) 2.3 3.3 q95 3.5 6.4 Pheat (MW) 11 31 frad 0.2 0.6 ¯ ne,core/nGW 0.68 0.96 ne,ped/nGW 0.5 0.75 ν∗ e,ped 0.9 2.0 βN1.0 2.4 Zeff 1.3 3.2 Sd5.1 5.9 cNe (%) <0.1 1.1 4. Achieved plasma parameters With the successful realisation of the QCE regime in JET with metal wall, a database with varying engineering and plasma parameters was assembled. This data set by no means reaches all the boundaries, but a natural scan in parameters within the given experimental time and machine constraints. Some of the main boundaries are specific technical limitations of JET, rather than physics limits of the QCE regime. For example, the maximum fuelling was restricted by a limit of the pressure inside the NBI duct. This limit depends on the plasma shape as well as SOL and far-SOL transport. We observed that when the close-to-double-null shape is reached, the duct pressure as well as the ICRF coupling resistance increase, a hint for an increased width of the edge density profile. The limitation on fuelling has a direct consequence on the achievable plasma current, with Ip=2.25 MA being the maximum in which QCE was observed in JET. A summary of achieved parameter ranges is presented in table 1. The values are averaged over stationary plasma conditions of at least 400 ms. 4.1. Pedestal top collisionality Achieving lower pedestal top collisionality as AUG and TCV was one of the goals of the JET experiments. We use the definition of [46] ν∗ e,ped =6.921 ×10−18 Zeff lnΛeRgeoq95ne,ped (ageo/Rgeo)1.5T2 e,ped (3) with the Coulomb logarithm lnΛe. A ν∗ e,ped ≈0.9–1.0 was achieved in various conditions, i.e. with both Ip=1.5 MA and 2.0 MA as well as with and without neon seeding (at both currents). Further reducing ν∗ e,ped proved difficult because of the limited amount of heating power (machine specific restriction) and the naturally high pedestal top density in the QCE regime. A detailed investigation on the pedestal in the QCE regime is left for future studies. 4.2. Safety factor The edge safety factor is an important parameter, both for the achievable fusion power in a future reactor as well as for the underlying physics of the QCE regime [18,21]. As mentioned above, more fuelling was needed to achieve QCE at higher plasma current and/or lower safety factor. The lowest safety factor q95 =3.5 was achieved with Ip=2.0 MA and |Btor|= 2.3 T, hence by reducing the toroidal field strength rather than increasing the plasma current. 7 Nucl. Fusion 65 (2025) 024003 M. Faitsch et al 5. Conclusions The QCE regime was successfully realised in JET in D and in D–T. This process was strongly assisted by the experimental findings and theoretical understanding in AUG and TCV that guided the approach in JET. The necessity of strong plasma shaping together with sufficient fuelling was confirmed in JET, in agreement with previous AUG and TCV results and with numerical predictions done for JET before the experiments [21]. A significant variation in the engineering parameters was achieved within the machine limits allowing further model verification. This QCE database will be further analysed in the future, with special emphasis on the separatrix conditions, the power fall-off length and the impact onto the main chamber wall, which all where shown to be important aspects of the QCE regime in both AUG [7,18,47] and TCV [15,24]. Acknowledgments This work has been carried out within the framework of the EUROfusion Consortium, partially funded by the European Union via the Euratom Research and Training Programme (Grant Agreement No. 101052200—EUROfusion). The Swiss contribution to this work has been funded by the Swiss State Secretariat for Education, Research and Innovation (SERI). Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union, the European Commission or SERI. Neither the European Union nor the European Commission nor SERI can be held responsible for them. Work supported in part by a grant PID2021-127727OB-I00 funded by the Spanish MCIN/AEI/10.13039/501100011033 and by ERDF ‘A way of making Europe’. ORCID iDs M. Faitsch https://orcid.org/0000-0002-9809-7490 M. Dunne https://orcid.org/0000-0002-5259-9970 I. Balboa https://orcid.org/0000-0002-5665-2222 P. Bilkova https://orcid.org/0000-0002-6156-9773 P. Bohm https://orcid.org/0000-0003-2590-4420 A. Kappatou https://orcid.org/0000-0003-3341-1909 D. Kos https://orcid.org/0000-0002-9550-4329 B. Labit https://orcid.org/0000-0002-0751-8182 O. Sauter https://orcid.org/0000-0002-0099-6675 S. Silburn https://orcid.org/0000-0002-3111-5113 E.R. Solano https://orcid.org/0000-0002-4815-3407 H.J. Sun https://orcid.org/0000-0003-0880-0013 A. Tookey https://orcid.org/0009-0009-6089-7634 E. Viezzer https://orcid.org/0000-0001-6419-6848 U. Stroth https://orcid.org/0000-0003-1104-2233 References [1] ITER Physics Basis Editors, ITER Physics Expert Group Chairs and Co-Chairs and ITER Joint Central Team and Physics Integration Unit 1999 Nucl. Fusion 39 2137 [2] Loarte A. et al 2014 Nucl. Fusion 54 033007 [3] Eich T., Sieglin B., Thornton A.J., Faitsch M., Kirk A., Herrmann A. and Suttrop W. 2017 Nucl. Mater. Energy 12 84 [4] Evans T.E. et al 2004 Phys. Rev. Lett. 92 235003 [5] Viezzer E. et al 2023 Nucl. Mater. Energy 34 101308 [6] Stroth U. et al 2022 Nucl. Fusion 62 042006 [7] Faitsch M., Eich T., Harrer G.F., Wolfrum E., Brida D., David P., Griener M. and Stroth U. 2021 Nucl. Mater. Energy 26 100890 [8] Stober J., Maraschek M., Conway G.D., Gruber O., Herrmann A., Sips A.C.C., Treutterer W. and Zohm H. (The ASDEX Upgrade Team) 2001 Nucl. Fusion 41 1123 [9] Saibene G. et al 2005 Nucl. Fusion 45 297 [10] Stober J. et al 2005 Nucl. Fusion 45 1213 [11] Perez von Thun C. et al 2008 Plasma Phys. Control. Fusion 50 065018 [12] Wolfrum E. et al 2011 Plasma Phys. Control. Fusion 53 085026 [13] Kirk A., Muller H.W., Wolfrum E., Meyer H., Herrmann A., Lunt T., Rohde V. and Tamain P. 2011 Plasma Phys. Control. Fusion 53 095008 [14] Harrer G. et al 2018 Nucl. Fusion 58 112001 [15] Labit B. et al 2019 Nucl. Fusion 59 086020 [16] Greenwald M. et al 1999 Phys. Plasmas 61943 [17] Griener M. et al 2020 Nucl. Mater. Energy 25 100854 [18] Faitsch M., Eich T., Harrer G.F., Wolfrum E., Brida D., David P., Dunne M., Gil L., Labit B. and Stroth U. 2023 Nucl. Fusion 63 076013 [19] Harrer G.F. et al 2022 Phys. Rev. Lett. 129 165001 [20] Radovanovic L., Dunne M., Wolfrum E., Harrer G., Faitsch M., Fischer R. and Aumayr F. 2022 Nucl. Fusion 62 086004 [21] Dunne M., Faitsch M., Radovanovic L. and Wolfrum E. 2024 Nucl. Fusion 64 124003 [22] Eich T. and Manz P. (The ASDEX Upgrade Team) 2021 Nucl. Fusion 61 086017 [23] Scott B.D. 2005 Phys. Plasmas 12 062314 [24] Stagni A. et al 2022 Nucl. Fusion 62 096031 [25] Matthews G.F. et al 2011 Phys. Scr. 2011 014001 [26] Gelfusa M. et al 2013 Rev. Sci. Instrum. 84 103508 [27] Campbell D.J. 2016 Europhys. News 47 28 [28] Philipps V., Mertens P., Matthews G.F. and Maier H. 2010 Fusion Eng. Des. 85 1581 [29] Balboa I. et al 2023 Plasma Phys. Control. Fusion 65 094002 [30] Bernard L., Helton F.J., Moore R.W. and Todd T.N. 1983 Nucl. Fusion 23 1475 [31] Dunne M.G. et al 2016 Plasma Phys. Control. Fusion 59 025010 [32] Bernert M. et al 2014 Plasma Phys. Control. Fusion 57 014038 [33] Huber A. et al 2017 Nucl. Mater. Energy 12 100 [34] Kukushkin A., Pacher H.D., Pacher G.W., Kotov V., Pitts R.A. and Reiter D. 2013 J. Nucl. Mater. 438 S203 [35] Goldston R., Reinke M.L. and Schwartz J.A. 2017 Plasma Phys. Control. Fusion 59 055015 [36] Pitts R. et al 2019 Nucl. Mater. Energy 20 100696 [37] Glöggler S. et al 2019 Nucl. Fusion 59 126031 [38] Giroud C. et al 2024 Nucl. Fusion 64 106062 8