Advances in the physics studies for the JT-60SA tokamak exploitation and research plan

Plasma Physics and Controlled Fusion PAPER Advances in the physics studies for the JT-60SA tokamak exploitation and research plan To cite this article: G Giruzzi et al 2020 Plasma Phys. Control. Fusion 62 014009 View the article online for updates and enhancements. Recent citations Conceptual design of JT-60SA edge Thomson scattering diagnostic R. Pasqualotto et al - Wall conditioning in fusion devices with superconducting coils T Wauters et al - This content was downloaded from IP address 2.140.173.162 on 29/06/2020 at 18:02 Advances in the physics studies for the JT-60SA tokamak exploitation and research plan G Giruzzi 1 , M Yoshida 2 , N Aiba 2 , J F Artaud 1 , J Ayllon-Guerola 3 , O Beeke 4 , A Bierwage 2 , T Bolzonella 5 , M Bonotto 5 , C Boulbe 6 , M Chernyshova 7 , S Coda 8 , R Coelho 9 , D Corona 9 , N Cruz 9 , S Davis 10 , C Day 11 , G De Tommasi 12,13 , M Dibon 14 , D Douai 1 , D Farina 15 , A Fassina 5 , B Faugeras 6 , L Figini 15 , M Fukumoto 2 , S Futatani 16 , K Galazka 7 , J Garcia 1 , M Garcia-Muñoz 3 , L Garzotti 17 , L Giudicotti 5 , N Hayashi 2 , M Honda 2 , K Hoshino 18 , A Iantchenko 8 , S Ide 2 , S Inoue 2 , A Isayama 2 , E Joffrin 1 , Y Kamada 2 , K Kamiya 2 , M Kashiwagi 2 , H Kawashima 2 , T Kobayashi 2 , A Kojima 2 , T Kurki-Suonio 19 , P Lang 14 , Ph Lauber 14 , E de la Luna 20 , G Marchiori 5 , G Matsunaga 2 , A Matsuyama 21 , M Mattei 12,13 , S Mazzi 22 , A Mele 12,13 , Y Miyata 2 , S Moriyama 2 , J Morales 1 , A Moro 15 , T Nakano 2 , R Neu 14 , S Nowak 15 , FP Orsitto 12 , V Ostuni 1,23 , N Oyama 2 , S Paméla 17 , R Pasqualotto 5 , B Pégourié 1 , E Perelli 15 , L Pigatto 5 , C Piron 24 , A Pironti 12,13 , P Platania 15 , B Ploeckl 14 , D Ricci 15 , M Romanelli 17 , G Rubino 25 , S Sakurai 2 , K Särkimäki 19 , M Scannapiego 11 , K Shinohara 2 , J Shiraishi 2 , S Soare 26 , C Sozzi 15 , T Suzuki 2 , Y Suzuki 27 , T Szepesi 28 , M Takechi 2 , K Tanaka 27 , H Tojo 2 , M Turnyanskiy 29 , H Urano 2 , M Valisa 5 , M Vallar 5 , J Varje 19 , J Vega 20 , F Villone 12,13 , T Wakatsuki 2 , T Wauters 30 , M Wischmeier 14 , S Yamoto 2 and R Zagórski 7 1 CEA, IRFM, F-13108 Saint-Paul-lez-Durance, France 2 National Institutes for Quantum and Radiological Science and Technology, Naka, Ibaraki, 311-0193, Japan 3 University of Seville, 41012 Seville, Spain 4 Rudolf Peierls Centre for Theoretical Physics, University of Oxford, United Kingdom 5 Consorzio RFX, I-35127 Padova, Italy 6 Université de Nice, Nice, France 7 Institute of Plasma Physics and Laser Microfusion, Hery 23, 01-497 Warsaw, Poland 8 EPFL, Swiss Plasma Center, CH-1015 Lausanne, Switzerland 9 Instituto de Plasmas e Fusão Nuclear, Instituto Superior Técnico, Universidade de Lisboa, P-1049-001 Lisboa, Portugal 10 Fusion for Energy, D-85748 Garching, Germany 11 Institute for Technical Physics, Karlsruhe Institute of Technology, Karlsruhe, Germany 12 Consorzio CREATE, Napoli, Italy 13 Università degli Studi di Napoli Federico II, Napoli, Italy 14 Max-Planck-Institut für Plasmaphysik, D-85748 Garching, Germany 15 Istituto di Fisica del Plasma CNR, 20125 Milano, Italy 16 Universitat Politècnica de Catalunya, Barcelona, Spain 17 CCFE, Culham Science Centre, Abingdon OX14 3DB, United Kingdom 18 Faculty of Science and Technology, Keio University, Yokohama, Kanagawa 223-8522, Japan 19 Aalto University, Department of Applied Physics, Aalto, Finland 20 Laboratorio Nacional de Fusion, CIEMAT, 28040, Madrid, Spain 21 National Institutes for Quantum and Radiological Science and Technology, Rokkasho, Aomori, 039- 3212, Japan 22 Aix-Marseille University, CNRS, PIIM, UMR, 7345, Marseille, France 23 Politecnico di Torino, Torino, Italy 24 EUROfusion /ENEA, Centro Ricerche Frascati, Italy Plasma Physics and Controlled Fusion Plasma Phys. Control. Fusion 62 (2020)014009 (16pp)https://doi.org/10.1088/1361-6587/ab4771 0741-3335/20/014009+16$33.00 © Gerardo Giruzzi Printed in the UK1 25 Università della Tuscia, Viterbo, Italy 26 Institute of Cryogenics and Isotopic Technologies—ICIT, 240050, Rm. Valcea, Romania 27 National Institute for Fusion Science, Toki, Gifu 509-5292, Japan 28 Wigner Research Centre for Physics, Budapest, Hungary 29 EUROfusion Programme Management Unit, D-85748 Garching, Germany 30 Laboratory for Plasma Physics, LPP-ERM/KMS, Brussels, Belgium E-mail: [email protected] Received 11 July 2019, revised 10 September 2019 Accepted for publication 24 September 2019 Published 24 October 2019 Abstract JT-60SA, the largest tokamak that will operate before ITER, has been designed and built jointly by Japan and Europe, and is due to start operation in 2020. Its main missions are to support ITER exploitation and to contribute to the demonstration fusion reactor machine and scenario design. Peculiar properties of JT-60SA are its capability to produce long-pulse, high-β,and highly shaped plasmas. The preparation of the JT-60SA Research Plan, plasma scenarios, and exploitation are producing physics results that are not only relevant to future JT-60SA experiments, but often constitute original contributions to plasma physics and fusion research. Results of this kind are presented in this paper, in particular in the areas of fast ion physics, highbeta plasma properties and control, and non-linear edge localised mode stability studies. Keywords: JT-60SA tokamak, high beta scenarios, fast ions, MHD (Some figures may appear in colour only in the online journal) 1. Introduction JT-60SA is a fully superconducting tokamak device jointly designed, built, and in the near future exploited by Japan and Europe under the Broader Approach Satellite Tokamak Programme, and under the Japanese national programme. It is the largest tokamak ever built before ITER and due to start operation in 2020 [1]. Both before and during the D-T phase of ITER, it will exploit and extend the legacy both of JET and of the superconducting tokamaks presently in operation (WEST, EAST, KSTAR). The main missions of JT-60SA are: (i)to support the ITER experimental programme as a satellite machine; (ii)at the same time, to pave the way to the next step of the international fusion programme, i.e., the demonstration fusion reactor (DEMO). More specifically, JT-60SA should provide key elements for the choice of DEMO parameters and the design of DEMO scenarios, in particular for a steadystate, advanced performance design option. In addition to the large size (allowing high current and energy confinement), JT-60SA has been conceived with the main capabilities required to accomplish these missions: high β(the ratio of kinetic to magnetic pressure), high shaping, long pulse, a high-power and flexible heating and current drive (H&CD) system, and dedicated magnetic coils to provide adequate control of scenarios close to the performance limits. The construction and assembly phase of the machine has been accompanied by the preparation of an extensive research plan [2], describing the various phases of the exploitation of the machine, in line with its progressive upgrades. This elaboration has included a number of coherent physics studies aimed at preparing a sound basis for the scientific exploitation of the machine [3]. They involve advanced modelling connected with the priorities of the scientific programme; conceptual studies of diagnostics and other sub-systems to improve the quality of the experiments; and development and validation of operation oriented tools. The preparation of the exploitation of JT-60SA has served as a catalyst leading to the production of physics results that are not only relevant to future JT-60SA experiments, but often constitute original contributions to plasma physics and fusion research, which are summarised and briefly described in this paper. The focus of these studies can be defined as the sustainment and control of high-βlong-pulse discharges in various plasma regimes. This requires a number of tools that have been studied and developed systematically: •global discharge simulation and integrated scenario modelling of its phases •wall conditioning (preparatory and inter-pulse) •breakdown and current ramp-up •flat-top sustainment and control (core and edge profiles, fast ions, power loads) •control and mitigation of instabilities and disruptions For each one of these elements, modelling studies have been carried out, but also specific projects of new machine subsystems have been launched and are being carried out by EU and Japanese teams for use in the commissioning and/or in the subsequent experimental campaigns. In some cases, preparatory experiments and tests have been carried out in EU devices. In this paper the ensemble of these studies and the contribution they give to the sustainment and control of high-βlong-pulse discharges in JT-60SA will be summarised, but specifichighlights will be given on the most recent and advanced results. 2 Plasma Phys. Control. Fusion 62 (2020)014009 G Giruzzi et al The paper is organized as follows: in section 2, the characteristics of the JT-60SA tokamak and the main elements of its experimental programme are briefly presented. In section 3, research advances towards controllable high-β, long-pulse scenarios in JT-60SA are described. Conclusions and prospects are presented in section 4. 2. The JT-60SA tokamak and experimental programme An extensive description of the JT-60SA device and actuator characteristics can be found in [1–4]and they are briefly summarised here. The machine has been conceived in order to attain high plasma performance and fully non-inductive discharges, not only through large size and high H&CD systems, but also through a high and flexible shaping capability. The main parameters are shown in table 1. Optimisation of the aspect ratio (A∼2.5)and shaping factor S=q 95 I p /(aB t ))∼7 (where I p is the plasma current, B t the toroidal magnetic field, a the minor radius, and q 95 the safety factor at 95% of the magnetic flux), in connection with the machine missions, have been key elements of the design, as well as the possibility of varying the plasma shape, from single null to double null, including an ITER-like shape. The machine’s characteristics allow performing discharges in the three main ITER regimes, i.e., H-mode at high current (5.5 MA), advanced inductive (hybrid), and steady state at lower current. A set of reference scenarios has been developed and is described in [2].Their main parameters are displayed in table 2. A non-inductive current drive (CD)is provided by the 10 MW negative-ion based neutral beam system (N-NBI)and by the 7 MW electron cyclotron (EC)wave system. Strong plasma heating by the 24 MW positive-ion based neutral beam system (P-NBI)is of course also an essential ingredient of the CD capabilities, via the bootstrap current associated to the pressure gradient. Part of the P-NBI power is injected tangentially, in the co-current (4MW)and in the counter-current (4MW)directions, allowing toroidal rotation control, which has a significant impact on plasma confinement and performance. The heating system also allows flexible sharing of the ion and electron heating, a crucial parameter for plasma transport properties. For power loads and density control, several generations of plasma facing components (PFCs)and in particular of divertors, are foreseen, with increasing capabilities of power flux (up to 15 MW m −2 ) and energy removal, eventually allowing full power discharges for 100 s current flat-top duration. The transition from carbon based to tungsten based PFC is foreseen after ∼10 years of machine operation. Control of plasma shape, position, error fields, and main magnetohydrodynamic (MHD)instabilities will be ensured by the characteristics of the equilibrium field coils, by an extensive set of in-vessel coils and by the electron cyclotron current drive (ECCD)system. In particular, control of the instabilities peculiar to high-βregimes, i.e., neoclassical tearing modes (NTMs)and resistive wall modes (RWMs)will be crucial for machine missions. Density control is essential for long pulse regimes and will be provided by a divertor cryopump system [5]for pumping, and for matter injection, by gas puff and pellet injection [6], which is also used for edge localised mode (ELM)pacing. An extensive set of diagnostics, for both control and scientific exploitation purposes, will be available and upgraded in the various phases of the exploitation of the machine, as described in appendix D of [2]. JT-60SA is now in an advanced phase of assembly, as shown by figure 1, and is due to start operation in 2020. The various research phases of the machine life are extensively discussed in [2]: the machine integrated commissioning will be carried out with hydrogen plasmas; this part of the initial research phase will also include experiments aiming at risk mitigation for the hydrogen phase of ITER. The following initial research phase in deuterium will explore the main ITER scenarios (H-mode and hybrid)and the advanced high-β scenario at progressively increasing heating power (up to 80% of the nominal power), but for relatively short pulses (∼5s), compatible with the initial C-coated divertor set. Installation of a new divertor made of carbon fibre reinforced monoblocks will allow the start of the integrated research phase, in which all the high-performance scenarios will be investigated for long pulses (∼100 s flat-top duration)and heating power close to the nominal value, using active cooling. Extensive studies of high-βsteady-state plasmas with optimum control strategies will be the main target of this phase. At this point, a major upgrade of the machine will be carried out, in which JT-60SA will become a fully metallic machine, with tungsten as the plasma facing material. High-performance scenarios combining high-density, high-β, little or no-ELMs, and a highly radiative divertor will be developed in this new configuration, in direct support of the contemporary ITER D-T experiments. Full exploitation of these scenarios (completely relevant to DEMO)will take place in the following extended research phase, in which further optimisation of the divertor shape will be attempted and the double-null configuration will be explored. Disruption management strategies, essential for Table 1. Nominal parameters of the JT-60SA tokamak. From top to bottom: toroidal magnetic field, plasma current, major and minor radii, aspect ratio, elongation and triangularity, plasma volume, flattop duration, total H&CD power, negative-ion based neutral beam power (and injection energy), positive-ion based neutral beam power (and injection energy), and EC heating power (and wave frequencies). B t 2.25 T I p 5.5 MA R/a 2.96/1.18 m A 2.5 κ/δ1.93/0.5 V p 133 m 3 t(flat-top)100 s H&CD power 41 MW N-NBI (500 keV)10 MW P-NBI (85 keV)24 MW ECRH (82, 110, 138 GHz)7MW 3 Plasma Phys. Control. Fusion 62 (2020)014009 G Giruzzi et al ITER, DEMO, and future tokamak reactors, will be developed and tested during the whole machine exploitation. 3. Research towards controllable high-β, long-pulse scenarios Systematic studies have been conducted with the main goal of preparing for the scientific exploitation of JT-60SA, but also for identifying the characteristics of diagnostics and actuators that should be developed in view of an efficient research programme. Discussion of the research results obtained can be organised around the typical sequence of items that have to be considered for the realisation of a successful high-β, longpulse discharge: from global discharge preparation via simulation, to wall conditioning, breakdown, ramp-up, flattop sustainment, MHD, and disruption control. Highlights of the main recent results in these areas are described in the following sub-sections. 3.1. Discharge simulation and integrated scenario modelling Predictive simulation of plasma discharges is the basis of the preparation of the experiment and the preliminary step for other analyses: MHD stability, sub-systems performance, and control strategies. Extensive scenario evaluation and preparation have been done using free-boundary equilibrium codes, such as ACCOME [7](including NBI sources)and TOSCA [8]. Global discharge simulations, producing time evolution of all the physical quantities, including magnetic equilibrium and radial profiles, are better done by fast simulators and indeed systematic analyses have been carried out with the METIS code [9]and reported in [2]. A much more accurate evaluation can be done if limited to particular phases Table 2. Main parameters of the JT-60SA reference scenarios. DN, SN: double-null, single-null configurations. I p : plasma current; B T : toroidal magnetic field; q 95 : safety factor at 95% of the poloidal magnetic flux; P add : additional heating power; P NNB : negative neutral beams power; P PNB : positive neutral beams power; P EC : EC power; n : eline-averaged electron density; f G : ratio of the line-averaged electron density to the Greenwald density; and β N : normalised plasma beta. #1#2#3#4-1 #4-2 #5-1 #5-2 #6 Inductive Inductive High density ITER-like Advanced inductive High β full-CD High β,f G full-CD 300 s high β Configuration DN SN SN SN SN SN SN SN I p (MA)5.5 5.5 5.5 4.6 3.5 2.3 2.1 2.0 B T (T)2.25 2.25 2.25 2.28 2.28 1.72 1.62 1.41 q 95 3.2 3 3 3.2 4.4 5.8 6 4 P add (MW)41 41 30 34 37 37 30 13.2 P NNB /P PNB /P EC 10/24/710/24/710/20/010/24/010/20/710/20/76/17/7 3.2/6/4 ne(10 19 m −3 )/f G 6.3/0.5 6.3/0.5 10/0.8 9.1//0.8 6.9/0.8 5.0/0.85 5.3/1.0 2.0/0.39 β N 3.1 3.1 2.6 2.8 3.0 4.3 4.3 3.0 Figure 1. Photograph of JT-60SA, showing the state of assembly in spring 2019. Reproduced with permission from Yutaka Kamada, Deputy Director General of the Naka Fusion Institute. 4 Plasma Phys. Control. Fusion 62 (2020)014009 G Giruzzi et al of the discharge, for instance the flat-top stationary phase, by integrated modelling codes, such as TOPICS [10], CRONOS [11], and JINTRAC [12]. In order to make such predictions as reliable as possible, a procedure for the validation of models and benchmark of integrated modelling codes has been employed, based on a set of reference JT-60U and JET discharges, representing the main scenarios (H-mode, hybrid, advanced). These discharges have been predictively simulated using both Japanese and EU codes, with a variety of transport, pedestal, rotation models, and scalings, with the aim of finding a unified modelling framework that works for the set of reference discharges of both machines, which are the most similar in size and characteristics to JT-60SA. In addition, this work has provided a benchmark of the integrated modelling suites of codes employed, with satisfactory results, as reported in [13–15]. Various transport models have been tested and can be used with comparable accuracy for predictive simulation of most of the reference discharges. In particular, the current diffusive ballooning mode (CDBM)heat transport model [16]provided accurate, or in some cases, conservative estimates of the electron and ion temperatures. A similar procedure is now being used for the current ramp-up phase [17]. On this basis, more complex phenomena can be analysed with the validated modelling framework, for instance, the formation of an internal transport barrier (ITB), its compatibility and alignment with non-inductively driven currents, and its controllability, which are basic ingredients of the steady-state high-βscenario 5 [2]. The CDBM transport model naturally allows ITB formation when the current density profile is peaked off-axis, as shown by the TOPICS simulation presented in figure 2. The strength of the ITB is found to be sensitive to the precise modelling of the N-NBI CD, in this case performed by means of the orbit-following Monte Carlo (OFMC)code [18]and including finite orbit width effects. Control of the ITB location and strength can be performed by small variations of the N-NBI power and/or by additional ECCD, and evaluated by means of this type of simulation [19]. Another phenomenon that is going to play a relevant role in integrated scenario modelling is plasma rotation, because of the impact of rotation shear on heat transport and generally plasma performance. JT-60SA has a flexible tool to control toroidal rotation, i.e., the torque due to co- and counterinjected P-NBI at 85 keV. However, the plasma response has to be evaluated taking into account braking due to non-axi- symmetric perturbations of the toroidal field, owing to ripple, error fields, and intentionally applied resonant magnetic perturbations (RMP)that will be used for ELM control. In JT- 60SA, these RMPs can be produced by the same system of coils that corrects error fields (error field correction coils (EFCCs)), aimed at suppressing locked modes. The ensemble of these perturbations give rise to the so-called neoclassical toroidal viscosity (NTV), an effect that is not usually included in integrated scenario modelling. This has been accomplished by an advanced modelling framework [20]combining in an iterative way: TOPICS [10]and OFMC [18]for the evaluation of the NBI induced torque, and the 3D neoclassical code FORTEC-3D [21]and the 3D equilibrium code VMEC [22]. An example of the toroidal rotation computed by means of this modelling framework for a high-current (5.5 MA) H-mode scenario of JT-60SA is shown in figure 3. In the left panel, red, green, and blue curves respectively display the profiles of toroidal rotation velocity without the NTV effect, with the NTV due to magnetic ripple and also including the effect of an n=3 RMP induced by 10 kA current in each of the EFCC, with even parity. In the right panel, the NTV torque (with and without RMP)is shown, together with the radial electric field. It is shown that significant breaking is caused by the ripple and should be taken into account in integrated modelling simulations. On the other hand, the n=3 RMP at 10 kA (which is typically used for ELM control)has a modest effect on the toroidal velocity, meaning that ELM control will not be detrimental to plasma rotation. However, higher applied currents in the EFCC (∼30 kA)will have a stronger impact and could be effectively used for rotation control, as discussed in [20]. 3.2. Wall conditioning, breakdown, and current ramp-up It is well known that plasma performance and discharge reproducibility in fusion machines strongly depend on the state of the walls, because of the impact of both main ion and impurity recycling. Recovery after disruptions usually needs particular care in order to have a broad enough operation window for breakdown. Because of the superconducting magnetic field, glow discharge cleaning will not be usable in JT-60SA between shots (as in ITER)and electron cyclotron wall conditioning (ECWC)is envisaged, a technique that has been tested but not fully validated yet, in particular at the 2nd EC harmonic. In order to improve the knowledge of this technique, dedicated experiments [23]have been performed on the TCV tokamak, demonstrating that after dedicated helium discharges with ECWC at the 2nd EC harmonic (extraordinary mode, X2), standard ohmic deuterium plasmas could then be sustained, whereas it would not have been possible without adequate wall conditioning. The discharge Figure 2. Simulations by the TOPICS code of the stationary phase of Scenario 5-1, but at reduced heating power (26 MW). Profiles of electron density (prescribed), current density, and ion and electron temperatures versus radial coordinate ρ(square root of normalised toroidal flux). 5 Plasma Phys. Control. Fusion 62 (2020)014009 G Giruzzi et al parameters on TCV were optimised to (i)minimise the absorption of EC stray radiation by in-vessel components by minimising the plasma breakdown time and maximising the absorption of power over the duration of the conditioning discharge, and to (ii)improve the discharge homogeneity and wall coverage, in particular towards the inboard surfaces where JT-60SA plasmas will be initiated. The main control parameters in this study are the addition of a poloidal magnetic field to the nominal toroidal field, the discharge pressure, and the EC power. In order to extrapolate these results to future JT-60SA ECWC experiments and routine use, modelling has been carried out by means of the TOMATOR-1D code [24]. This one-dimensional reactiondiffusion-convection code has been developed to simulate plasma production by radio-frequency waves inside a tokamak, using the Braginskii continuity and heat balance equations. In this algorithm, the evolution of the radial density and temperature profiles is documented for nine species for H 2 /He/C plasma mixtures from a transient to a steady-state condition. The plasma simulator is used to study EC absorption and transport properties as a function of the applied vertical magnetic field component. Figure 4shows the experimental electron density measured by Thomson scattering on TCV at w w=2ce,and the simulated density by TOMATOR-1D for X2-ECRH plasma at 1.5 T, 400 kW of launched power, toroidal, and poloidal injection angles of 19.3°and 7.0°, respectively. The simulations, reproducing the experimental data, predict Bohm-like diffusion transport, convection losses of the order of 10–100 m s −1 , and an EC absorption scaling proportional to µPP nT abs in e e /as predicted by the quasi-optical beam-tracing code GRAY [25]. The dependence of transport processes on the vessel dimensions via a multi-machine study is required to arrive at predictive capabilities for ECWC plasma on JT-60SA. Breakdown and plasma initiation is a particularly delicate phase in a new machine, operating with a new control system. Assistance by EC waves for pre-ionisation and plasma heating to burnthrough can prove essential to improving the efficiency of this task and will be mandatory in ITER [26]. Predictive modelling of EC-assisted breakdown could significantly speed up the search for optimised breakdown configurations, however it requires a complex simulation framework, combining free-boundary equilibrium equations, balance equations for the time evolution of energy, particles (with ionisation/recombination terms and impurities)together with the circuit equation for the plasma current, and EC wave propagation equations, including at least the first wall reflection. To this end, a dedicated modelling workflow has Figure 3. Simulation by means of an integrated modelling framework including the codes TOPICS, OFMC, FORTEC-3D, and VMEC for H-mode high current and the high-density Scenario 3. Left: deuterium ion toroidal velocity profile; all the curves correspond to simulations with the NBI torque; the green curve also includes NTV and the blue curve NTV and RMP (n=3, 10 kA current in each of the EFCC coils). Right: NTV torque, with and without RMP (n=3, 10 kA), and radial electric field. Figure 4. Electron density measured by Thomson scattering and the same quantity simulated by TOMATOR-1D versus vertical magnetic field (in percentage of toroidal field)at ww=2ce,for X2-ECRF plasma on TCV at B t =1.5 T, 400 kW of launched power. 6 Plasma Phys. Control. Fusion 62 (2020)014009 G Giruzzi et al been developed and is being validated on experiments [27].It combines the breakdown code BKD0 [28], the beam-tracing code GRAY [25]and CREATE-BD, a specific version of the free-boundary equilibrium code CREATE [29], which integrates and optimises the active circuit currents, taking into account eddy currents in the passive structures for developing the plasma breakdown scenario. Furthermore, the BKD0 impurity model, consistent with the code DYON [30]for carbon wall machines, has been validated on TCV experiments. Several cases have been treated, for the two main frequencies of the JT-60SA gyrotrons: 110 GHz and 138 GHz [31], the additional operation frequency of 82 GHz, and various injection angles. The electron temperature and plasma current depend on ECRH power and present a threshold for successful start-up. Figure 5shows this computed EC power threshold as a function of the initial neutral H 2 pressure for two cases: (i)138 GHz (2nd harmonic)X-mode, injection perpendicular to the magnetic field and at fixed poloidal angle 35.5°with respect to the horizontal direction (blue curve)and (ii)82 GHz (1st harmonic)O-mode, optimised injection angles (green), i.e., poloidal angle 21°and toroidal angle (with respect to the perpendicular)of 20°. For this second case, two different values of the initial C content and oxygen fraction n O /n H =0.1% have been considered. As the initial pressure increases, more power is required to overcome the radiation barrier. The steerable launcher that will equip JT- 60SA after the commissioning phase will allow optimising the absorption and increase the operational space even in the presence of impurities: the main difference between the two cases is related to the fact that injection angle optimisation by means of a steerable launcher allows for a second pass after the first reflection through the initial small plasma, with the change of polarisation from the O- to X-mode, which greatly increases the EC wave effect owing to substantial increase in the second pass absorption. In order to monitor this delicate phase (as well as for general plasma overview)a wide-angle version of the EDICAM visible camera [32]has been designed and manufactured as a direct European contribution to JT-60SA diagnostics. It will be an important tool for the detection of dangerous events, such as hot-spots, plasma boundary identification with a temporal resolution up to 1 kHz (comparable to magnetic equilibrium reconstruction), disruptions, scrapeoff layer (SOL)statistical properties etc. Moreover, this will be the phase in which the machine could be more likely prone to risks connected with EC stray radiation, owing to incomplete absorption of the EC wave beams. This has motivated specific studies [33,34], in which an analysis of the residual non-absorbed ECRF power fraction expected in the various applications and plasma scenarios has been carried out, studying its dependence on the steering angle and on the plasma’s main parameters such temperature and density. Both transient conditions, such as plasma start-up and flat-top scenarios have been taken into consideration. Moreover, the expected stray power density distribution in the vessel and particularly around the potentially critical areas such as diagnostics windows or pumping ducts has been evaluated. An adequate detection system to limit the risks related to stray radiation is presently considered. After breakdown and burnthrough are obtained, the current and density ramp-up is a crucial phase for the establishment of high-performance scenarios while saving central solenoid flux: X-point formation, high shaping control, H-mode transition, and in particular the access to the safety factor profiles typical of JT-60SA advanced regimes (hybrid and steady state). Earlier simulations of this phase have been performed with the MECS code [35,36]and benchmarked [37]with the CREATE-NL code [38]. A way to improve the accuracy of this type of simulation is to couple a freeboundary equilibrium code with a plasma model including the main ingredients of integrated modelling, i.e., heat and particle transport, current diffusion, and H&CD modules. Although some examples of such a combination of codes exist in the literature, the required computation time is usually extremely large. In order to keep it at a level that makes this type of modelling practical and useful for preparing experimental scenarios, the fast integrated modelling code METIS [9]is used as plasma module and has been coupled to two different free-boundary equilibrium codes, for alternative uses: CREATE-NL [38]and FEEQS [39]. FEEQS is used in inverse mode in conjunction with METIS to compute the premagnetisation phase and to initialise and optimise the plasma scenario, verifying the coil limits. An example is given in figure 6, showing three snapshots of free-boundary equilibria computed by FEEQS (top)and the self-consistent evolutions of the poloidal beta β pol , internal inductance l i , electron temperature, and current density profiles computed by METIS (bottom). A full simulation including controllers can be performed by means of CREATE-NL exchanging equilibrium and profiles data with METIS in the framework of a SIMULINK scheme. Figure 5. EC-assisted breakdown simulation with BKD0, GRAY, and CREATE-BD. EC power threshold for successful breakdown as a function of the initial neutral H 2 pressure for 1st harmonic O-mode (optimised injection angles and two passes within the breakdown region)and 2nd harmonic X-mode (fixed angles, single pass),at different initial C content and n O /n H =0.1%. The red dot represents the ohmic breakdown case. 7 Plasma Phys. Control. Fusion 62 (2020)014009 G Giruzzi et al 3.3. Flat-top plasma sustainment: density, power loads, and radiation Density control is a key ingredient for long pulse regimes and in particular for access to high-density scenarios, close to the Greenwald limit. The two main actuators of such control will be the divertor cryopump system [5]and the pellet fuelling system, which is being designed [6]. This will be an advanced system, conceived for combined fuelling and ELM pacing using two extruders. Detailed modelling of pellet ablation, using the HPI2 code [40], has been carried out in order to determine the optimum pellet injection location and parameters for all the reference scenarios and optimise the system design. The same code has been used in conjunction with the integrated modelling suite JINTRAC, appropriately set up for JT-60SA simulations [15], in order to assess the feasibility of combined fuelling and ELM pacing. The JINTRAC simulation starts from parameters typical of Scenario 2 (see table 2) and models the growth of the density due to the injection of trains of pellets with frequency 12.5, 11.1, 10.0, 9.1, 8.3, and 7.7 Hz, respectively. The pellets are injected every 5 s, which allows attaining nearly stationary densities. In addition, ELM pacing pellets at 50 Hz are injected between fuelling pellets. This corresponds to the injection patterns of two centrifuge arms rotating at 50 Hz with a fuelling pellet injected every 4, 4.5, 5, 5.5, 6, and 6.5 rotations respectively. The pellet injection velocity was set to 400 m s −1 and the masses of the fuelling and ELM pacing pellets are 6.5 10 20 atoms and 0.8 10 20 atoms, respectively. The results for the case with fuelling pellets only are shown in figure 7. The electron density profile versus time and normalised radius is shown in the top panel and the electron temperature in the bottom panel. It can be seen that it is possible to increase the density from the level characteristic of Scenario 2 to significantly higher density (typical of Scenario 3)within a few seconds. After the fast initial transient, a slow decrease of the average density is also seen as the pellet injection frequency is gradually reduced. In the corresponding case with simultaneous fuelling and ELM pacing pellets, there is an additional contribution by the pacing pellets that increases the density by 25%–30% [6],as shown in table 3. This contributes significantly to the fuelling of high-density scenarios and should be taken into account in the global control strategy of the discharge. The second key element for long pulse regimes is the control of power loads on the divertor during the strong heating phase, which can be attained by divertor radiation associated with controlled impurity seeding strategies. Systematic studies of the relevant scenarios, but in a simplified geometry, have been performed by means of the COREDIV code, both for C-PFC [41–43]and W-PFC [42,44]. Simulations combining the core, SOL, and divertor have been performed for a scenario similar to Scenario 5-1, but at reduced heating power, using JINTRAC coupled to the edge code EDGE2D and the Monte Carlo code for the neutrals EIRENE [45]. A comparison of COREDIV and EDGE2D carbon radiation patterns for Scenario 3 (high density)are Figure 6. Results of simulations for Scenario 2 (H-mode at plasma current 5.5 MA)combining FEEQS and METIS. Top: snapshots of freeboundary equilibria at three different times, from left to right: in the initial phase of the current ramp-up (t=0.6 s, I p =0.5 MA), during the ramp-up (t=5s,I p =3MA)and in the flat-top phase (t=50 s, I p =5.5 MA). Bottom: time evolution of β pol and l i ; electron temperature and current density profiles at different times. 8 Plasma Phys. Control. Fusion 62 (2020)014009 G Giruzzi et al 4. Conclusions and prospects The JT-60SA scientific programme is both broad and ambitious, as appropriate to a large and international experimental device that will be at the forefront of the international fusion programme for many years. The physics studies presented in this paper, although not an exhaustive description, provide relevant examples of the efforts necessary to build a physics basis for realising the main research objectives defined in the research plan [2]. Such studies will accompany the machine enhancement programme that will be developed in parallel with the operation and scientific exploitation phase. The most significant enhancement will be the transition to W-PFC, for which R&D is in progress, including tests of specifically developed W coatings [80]. Ideas of advanced diagnostics are being developed for possible application in the integrated and extended research phases. For instance, in connection with the key objective of current profile control, conceptual design of a polarimetry system has been carried out [81]. In connection with turbulence studies, as, for instance, those reported in [82], turbulence diagnostics for both the core and edge have been proposed and are being designed [83,84]. The forthcoming start of the machine operation (2020)is expected to further intensify these prospective studies. Acknowledgements This work has been carried out within the framework of the EUROfusion Consortium and has received funding from the EURATOM research and training programme 2014–2018 and 2019–2020 under grant agreement no. 633053. The views and opinions expressed herein do not necessarily reflect those of the European Commission. The authors gratefully acknowledge members of the JT-60SA Integrated Project Team for data exchange and fruitful discussions. ORCID iDs G Giruzzi https://orcid.org/0000-0001-9628-5968 References [1]Barabaschi P et al 2019 Nucl. Fusion 59 112005 [2]JT-60SA Research Unit, ‘JT-60SA Research Plan. Research Objectives and Strategy’, v4.0, Sept. 2018 (http://jt60sa. org/pdfs/JT-60SA_Res_Plan.pdf) [3]Giruzzi G et al 2017 Nucl. Fusion 57 085001 Figure 16. Disruption simulation by the CarMa0NL code. Plasma equilibrium configurations during a disruption (downwards VDE); halo currents are shown in green. Left: time at which the plasma hits the wall; right: configuration 5 ms after the start of the CQ. The maximum halo current reaches about 25% of the initial plasma current. 15 Plasma Phys. Control. Fusion 62 (2020)014009 G Giruzzi et al [4]Kamada Y et al 2011 Nucl. Fusion 51 073011 [5]Luo X et al 2018 Fusion Eng. Des. 136 467 [6]Lang P et al 2019 Fusion Sci. Technol. 75 178 [7]Tani K et al 1992 J. Comp. Phys. 98 332 [8]Shinya K 2000 J. Plasma Fusion Res. 76 479 [9]Artaud J-F et al 2018 Nucl. Fusion 58 105001 [10]Hayashi N et al 2010 Phys. Plasmas 17 056112 [11]Artaud J-F et al 2010 Nucl. Fusion 50 043001 [12]Romanelli M et al 2014 Plasma Fusion Res. 93403023 [13]Garcia J et al 2014 Nucl. Fusion 54 093010 [14]Hayashi N et al 2017 Nucl. Fusion 57 126037 [15]Garzotti L et al 2018 Nucl. Fusion 58 026029 [16]Honda M and Fukuyama A 2006 Nucl. Fusion 46 580 [17]Morales J et al 2019 Proc. 46th EPS Conf. Plasma Phys. (Milano)P5.1011 [18]Tani K et al 1981 J. Phys. Soc. Jpn. 50 1726 [19]Hayashi N et al 2018 27th IAEA Fusion Energy Conf. (Ahmedabad, India, 2018)PPC/1-1 [20]Honda M et al 2018 Nucl. Fusion 58 112012 [21]Satake S et al 2011 Plasma Phys. Control. Fusion 53 054018 [22]Hirshman S P et al 1986 Comput. Phys. Commun. 43 143 [23]Douai D et al 2018 Nucl. Fusion 58 026018 [24]Wauters T et al 2019 46th EPS Conf. Plasma Phys. (Milano)I. 102, to be unpublished to Plasma Phys. Control. Fusion [25]Farina D 2007 Fusion Sci. Technol. 52 154 [26]Stober J 2011 Nucl. Fusion 51 083031 [27]Ricci D et al 2019 Proc. 46th EPS Conf. Plasma Phys. (Milano)O5.103 [28]Granucci G et al 2015 Nucl. Fusion 55 093025 [29]Albanese R and Villone F 1998 Nucl. Fusion 38 723 [30]Kim H-T et al 2012 Nucl. Fusion 52 103016 [31]Kobayashi T et al 2015 Fus. Eng. Des. 96–97 503 [32]Zoletnik S et al 2013 Fus. Eng. Des. 88 1405 [33]Moro A et al 2017 Fusion Eng. Des. 123 435 [34]Sozzi C et al 2019 Proc. 46th EPS Conf. Plasma Phys. (Milano)P4, 1089 [35]Urano H et al 2015 Fus. Eng. Des. 100 345 [36]Miyata Y et al 2014 Plasma Fus. Res. 93403045 [37]Cruz N et al 2017 Control Eng. Pract. 63 81 [38]Albanese R et al 2015 Fus. Eng. Des. 96–97 664 [39]Heumann H et al 2015 J. Plasma Phys. 81 905810301 [40]Pégourié B et al 2007 Nucl. Fusion 47 44 [41]Zagórski R et al 2016 Nucl. Fusion 56 016018 [42]Zagórski R et al 2017 Nucl. Fusion 57 066035 [43]Gałązka K et al 2018 Contrib. Plasma Phys. 58 751 [44]Gałązka K et al 2017 Plasma Phys. Control. Fusion 59 045011 [45]Romanelli M et al 2017 Nucl. Fusion 57 116010 [46]Kawashima H et al 2016 Contrib. Plasma Phys. 56 778 [47]Hoshino K et al 2018 Contrib. Plasma Phys. 58 638 [48]Wiesen S et al 2015 J. Nucl. Mater. 463 480 [49]Rubino G et al 2019 Proc. 46th EPS Conf. Plasma Phys. (Milano)P5.1039 [50]Chernyshova M et al 2019 Proc. 46th EPS Conf. Plasma Phys. (Milano)P1.1012 [51]Vallar M et al 2019 Proc. 46th EPS Conf. Plasma Phys. (Milano)P5.1008 [52]Varje J et al 2019 Proc. 46th EPS Conf. Plasma Phys. (Milano) P2.1086 [53]Bierwage A et al 2019 Plasma Phys. Control. Fusion 61 014025 [54]Coelho R et al 2019 Proc. 46th EPS Conf. Plasma Phys. (Milano)P4.1031 [55]Garcia J et al 2015 Nucl. Fusion 55 053007 [56]Ayllon-Guerola J et al 2019 Proc. 46th EPS Conf. Plasma Phys. (Milano)P1.1009 [57]Perelli-Cippo E et al 2019 Proc. 46th EPS Conf. Plasma Phys. (Milano)P2.1011 [58]Lauber P et al 2007 J. Comput. Phys. 226 447 [59]Lauber P 2013 Phys. Rep. 533 33 [60]Borba D et al 2004 Plasma Phys. Control. Fusion 46 809 [61]Gassner T et al 2012 Phys. Plasmas 19 032115 [62]Girardo J-B et al 2016 Phys. Plasmas 23 012505 [63]Vega J et al 2019 Fus. Eng. Des. 146 1291 [64]Dibon M et al 2019 Proc. 46th EPS Conf. Plasma Phys. (Milano)P5.1006 [65]Huysmans G and Czarny O 2007 Nuclear Fusion 47 659 [66]Pamela S et al 2018 27th IAEA Fusion Energy Conf. (Ahmedabad, India, 2018)OV/4-4 [67]Aiba N et al 2018 Plasma Phys. Control. Fusion 60 014032 [68]Futatani S et al 2014 Nucl. Fusion 54 073008 [69]Gal K et al 2008 Nucl. Fusion 48 085005 [70]Pigatto L et al 2019 Nucl. Fusion 59 106028 [71]Albanese R et al 2008 IEEE Trans. Magn. 44 1654 [72]Pigatto L et al 2019 Proc. 46th EPS Conf. Plasma Phys. (Milano)P5, 1002 [73]Shiraishi J et al 2017 Nucl. Fusion 57 126051 [74]Sauter O et al 2002 Plasma Phys. Control. Fusion 44 1999 [75]Ramponi G et al 1999 Phys. Plasmas 63561 [76]Basiuk V et al 2017 Plasma Phys. Control. Fusion 59 125012 [77]Nowak S et al 2019 Submitted for publication [78]Villone F et al 2013 Plasma Phys. Control. Fusion 55 095008 [79]De Tommasi G et al 2018 27th IAEA Fusion Energy Conf. (Ahmedabad, India, 2018)EX/P3-26 [80]Fukumoto M et al 2017 Phys. Scr. 2017 014029 [81]Boboc A et al 2018 Plasma Phys. Control. Fusion 60 075016 [82]Beeke O et al 2019 Proc. 46th EPS Conf. Plasma Phys. (Milano)P5, 1013 [83]Coda S et al 2019 Proc. 46th EPS Conf. Plasma Phys. (Milano)P1, 1013 [84]Asztalos Ö et al 2017 Fusion Eng. Des. 123 861 16 Plasma Phys. Control. Fusion 62 (2020)014009 G Giruzzi et al