Session CO07: Magnetic Confinement: Research in Support of ITER

Live

SPI cryogenic impurity injectors have been fabricated and installed for use in disruption mitigation and runaway electron dissipation experiments on DIII-D, JET and KSTAR. These systems are now all operational and being utilized in disruption mitigation research in support of ITER and are based on a 3-barrel design for flexible pellet size and variable pellet composition studies. We will present performance results of these SPI systems and describe how the systems are being used in experiments to study radiation symmetry during the thermal quench and forces encountered during the current quench as well as the interaction of injected material with runaway electron beams that may form during the current quench as a means of mitigating possible impact damage from the energetic electrons. Physics and operational lessons learned on propellant gas removal, multi-pellet synchronization, and radiation asymmetry using this technology and the impact on the ITER SPI based disruption mitigation system design will be described.   

On Demand

To validate the disruption mitigation strategy of ITER discharge with multiple injections of the Shattered Pellet Injection (SPI) system, two identical SPI systems were installed symmetrically at the opposite position on the toroidal plane and they have been tested on KSTAR discharges. Preliminary test results indicated that synchronicity in the arrival time of the two pellets is critical among other factors in increasing the success rate of the current quench thus increasing radiation dissipation$^{\mathrm{\ast }}$. Many diagnostic systems such as MHD spectroscopy, CES, and ECE are employed to support the DECAF analysis. The real-time data from these diagnostics and the state-of-the-art ECEi system (i.e., 2-D electron temperature) as well as MSE will be utilized for real-time DECAF analysis for the first time. Then the RT DECAF analysis will be integrated into the overall KSTAR control system so that the disruption avoidance system can be fully tested.$^{\mathrm{\ast \ast }}$ In this paper, the latest status of the SPI experiment and the integration of the advanced diagnostics for RT DEDCAF capability will be reported.   

Live

This work provides a qualitative description of the pellet response to the ambient runaway electrons. For ITER-relevant parameters, our estimates suggest that the cryogenic pellets will be sublimated instantly at the edge of the runaway beam. The subsequent rapid expansion of the sublimated material spreads the impurities over the poloidal cross-section of a tokamak on a millisecond time scale prior to the complete ionization of the expanding cloud. The injected solid pellet turns into a rapidly expanding gas cloud before it reaches the core of the runaway beam. As a result, the pellet acts similar to the massive gas injection.   

Live

Power exhaust solutions for a next-step device like ITER must be compatible with high plasma performance. In particular, a high separatrix density is necessary to achieve sufficiently low divertor power loading and type-I ELMs must be avoided or mitigated. Recent optimisation at ASDEX Upgrade has led to a regime highly suitable for power exhaust, combining high core performance with high separatrix density at high triangularity, close-to-double-null as foreseen for ITER. It is characterised by enhanced filamentary transport preventing type-I ELMs and leading to a quasi-continuous exhaust, with the interaction between filaments and divertor detachment still being an open point of research. While for type-I ELMs the whole pedestal is affected, here only the pedestal foot is altered. In this region nowadays machines match the collisionality for ITER in power exhaust compatible regimes. The most crucial power exhaust parameter, the power fall-off length, is shown to widen up to a factor of four w.r.t. the ITPA-multi-machine scaling.   

Live

The erosion of neutral tungsten at the plasma boundary is diagnosed in the Compact Toroidal Hybrid (CTH) experiment with a high-resolution spectrometer and a new Langmuirprobe allowing for simultaneous electron temperature, density and W spectral measurements. This data is used to constrain collisional-radiative modeling with ColRadPy, which suggests that neutral tungsten emission is dominated by non-steady-state metastable populations over a wide range of plasma conditions. Inferred non-steady-state metastable populations are presented using W I spectral lines near 260 nm. High-resolution tungsten spectra from CTH plasmas are compared to modeled emission using W I R-matrix excitation data and exchange classical impact parameter (ECIP) ionization constrained by Teand nefrom Langmuir probe measurements. The addition of a Chodura sheath for modeling fusion relevant divertor conditions and CTH plasmas suggest changes of the W relative metastable fraction by up to 50{\%} when compared to modeling with no sheath, underlining the importance of understanding metastable effects when diagnosing W emission in fusion edge plasmas.   

Live

We report on a theoretical discovery of new regimes of Alfvén eigenmode (AE) induced fast ion transport in tokamak plasmas, where microturbulence plays the role of a mediator of fast ion relaxation. Coulomb collisional scattering alone leads to relatively small AE amplitudes in ITER-like plasma conditions and does not reproduce the steady state regimes observed in present day experiments in conventional tokamaks. We show that in nonlinear regimes the effective pitch angle scattering due to microturbulence can lead to steady state AEs with up to an order of magnitude higher amplitudes since the microturbulence scattering is expected to be 2-5 times stronger than the Coulomb scattering. This indicates a new route for fast ion radial transport and its intermittency, which are beyond the scenarios described in “Energetic ion transport by microturbulence is insignificant in tokamaks” [D. C. Pace et al., Phys. Plasmas 20 (2013) 056108]. The increase of AE amplitudes due to microturbulence in predictive simulations for burning plasmas was ignored earlier but needs to be accounted for when designing future plasma scenarios.   

Live

New experimental results from ASDEX Upgrade discharges in H and D are presented and applied to the validation of the quasi-linear turbulent transport models TGLF and QuaLiKiz. Linear gyrokinetic calculations were performed for verification. The dataset comprises ECRH power scans from the low to the intermediate density range, covering the turbulent transport conditions expected during the ITER PFPO-1 phase. For the success of the initial phase of ITER operation, reliable predictions of the central electron temperature (for X3 ECRH absorption) and of the ion heat flux at the periphery (for H-mode access) are essential. These are completely determined by the electron and ion turbulent heat transport in the core. While TGLF accurately reproduces the measured central electron temperature and the ion heat flux at the edge, QuaLiKiz tends to overpredict central T$_{\mathrm{e}}$ in TEM-dominated discharges. The source for this discrepancy is currently under investigation.   

Live

The ITER Integrated Modelling {\&} Analysis Suite (IMAS) has been developed to provide a standard and modular framework for supporting scenario preparation and plasma operation through a standardised data model designed to support both simulated and experimental data [1]. One of the most sophisticated physics workflows developed so far in IMAS is for Heating and Current Drive (H{\&}CD) modelling. The IMAS Python H{\&}CD workflow has been developed by the ITER Organization based upon earlier developments carried out within the EU [2]. It can be coupled to any transport solver adapted to IMAS as a natural consequence of using IMAS's standard Interface Data Structures. It can simulate the synergy between H{\&}CD sources and provides a high degree of modularity between various H{\&}CD models including all the heating sources available in ITER, i.e. ECRH, ICRH, NBI and fusion reactions. This enables it to describe all scenarios in the ITER Research Plan. In this work, the modelling results of the H{\&}CD workflow will be presented for the Pre-Fusion Power Operation (PFPO) phase of the ITER Research Plan. [1] F. Imbeaux et al, Nucl. Fusion 55 (2015) 123006 [2] G. Falchetto et al, 26$^{\mathrm{th}}$ IAEA FEC, Kyoto, Japan (2016)   

Live

Here we study the possibility for ITER ELMy H-mode operation in hydrogen plasma at 7.5MA/2.65T using up to 33MW NBI and 20 -- 30MW ECH, intended for commissioning of the ELM control systems. Two strategies are considered to lower $P_\textrm{L--H}$: a) operation at low density, or b) introducing $\sim$10\% He (since this is observed to decrease $P_\textrm{L--H}$ in JET~[1]). The scenarios are modelled self-consistently using JINTRAC~[2], developed by EUROfusion for integrated core (JETTO/SANCO/EDWM), edge and SOL/divertor modelling (EDGE2D/EIRENE). Results obtained so far indicate that stationary ELMy H-mode is indeed accessible, with and without 10\% He, provided that the ECH power level is 30MW. [1] D.C. McDonald, \emph{et al.}, EFDA-JET Report No.~EFDA--JET--CP(10)08/24 (2011) [2] M. Romanelli, \emph{et al.}, \emph{Plasma and Fusion Research} {\bf 9}, 3403023 (2014)   

Live

Impurity transport is studied in low torque ITER relevant plasmas by varying the injected torque and plasma rotation via neutral beam injection at fixed input beam and electron cyclotron (EC) power, while other plasma parameters are held nearly constant. Trace amounts of aluminum and tungsten are injected with a laser blow-off system at three injected torque levels: 3, 1.5, and $\sim$ 0 N-m. As the core toroidal rotation decreased by 5x over this scan, the core ion temperature decreased by 25\%. EC power is applied to study the effect of EC location on impurity transport. The W impurity confinement time of the 3 N-m plasmas is $\sim$ 400 ms and it increased to $\sim$ 750 ms at $\sim$ 0 N-m; the core $Z_{eff}$ also increased from 1.9 to 2.7. Similarly, the confinement time for Al increased from 200 to 400 ms over the same torque range. Preliminary analysis indicates Al core transport coefficients are nearly the same at low and high torque, the increase in impurity confinement times and $Z_{eff}$ are likely due to a drop in ELM frequency. Understanding the relationship between rotation, ELM frequency, EC heating, and impurity mass has important implications for ITER and will be further investigated via impurity transport modeling and experimental turbulence measurements.   

Live

An assessment of ITER steady-state plasma operation using 0-D operational space scans and ASTRA/KINX ideal MHD stability analysis has shown that Q\textasciitilde 5 fully non-inductive operation can be achieved by using only the ITER NBI and EC systems in the range of foreseen upgrades [A.R. Polevoi, et al., 46th EPS Conference, 8-12.07.2019, ECA Vol. 43C, P5.101]. This analysis revealed that operating the plasma at 9-10MA and 60-80{\%} Greenwald density would be necessary to achieve a sufficient current-drive efficiency that maintains the required fusion power gain. However, an optimization of NBI and off-axis ECCD will be essential to find a target plasma state that avoids core MHD instabilities by tailoring the current profile and relaxing the requirement on the energy confinement enhancement. An upgrade of NBI power up to 49.5MW by adding a 3rd beam line and the use off-axis ECCD with a power level of 20-30MW from the equatorial and upper launchers are foreseen. In the present study, access to fully non-inductive Q\textasciitilde 5 ITER steady-state operation with NBI and EC heating and current drive has been modelled to demonstrate the Q\textasciitilde 5 steady-state goal with the NBI and EC power upgrade options and within CS/PF coil limits, provided that the desired energy confinement (H98$=$1.5-1.6) can be reached. Integrated scenario simulations using CORSICA show that ITER would be able to develop a strategy for profile tailing during the current ramp and accessing the target steady-state plasma.   

On Demand

Steady-state high performance plasma up to 60s has been demonstrated on EAST with dominant electron heating ($T_{\mathrm{e}}=$ 5keV, $f_{\mathrm{GW}}=$ 0.78, $H_{\mathrm{98,y2}}=$ 1.3) by using zero torque radio-frequency (RF) heating (LHW$+$ECRH). Meanwhile, extending to high fusion performance has been explored by applying moderate (neutral beam) NB power ($\beta_{\mathrm{N}}=$ 2.0, $\beta_{\mathrm{p}}=$ 2.5, $H_{\mathrm{98,y2}}=$ 1.2). Synergy effect of electron internal transport barrier and Shafranov shift stabilizing of turbulence improves the energy confinement in this scenario. Efforts towards high fusion gain towards ITER baseline have been made at lower $q_{\mathrm{95\thinspace }}$and torque. Recent experiments achieved $\beta_{\mathrm{N}}=$ 1.55 at $q_{\mathrm{95}}=$ 3.2 and torque $T_{\mathrm{inj}}=$ 0.33Nm with the upper ITER-like tungsten divertor, which was obtained by utilizing lower hybrid wave heating and current drive along with NB power. The $n=$ 4 resonant magnetic perturbation coil was applied as an integrated control technique of ELM mitigation ($f_{\mathrm{ELM}}=$ 400-600Hz) and high-Z impurity exhaust, while had little degradation impact on plasma performance. In the plasma core region, sawtooth oscillation was observed by soft X-ray diagnostic, suggesting the existing of $q$0 \textless 1.0.   

Live

Results from coordinated research on DIII-D and EAST are illustrating the promise of high poloidal-beta ($\beta_{\mathrm{P}})$ tokamaks for attractive fusion power reactors. By optimizing at low plasma current and high plasma pressure, high-$\beta_{\mathrm{P}}$ operation drastically reduces the disruptivity and potential disruption damage, the requirements on external current drive, the ELM size and ELM control challenge, and the difficulty of divertor detachment, while a high energy confinement time (despite the low plasma current) is achieved through Shafranov shift suppression of turbulence enhanced by core density gradients. Fully noninductive operation with a tungsten divertor has been demonstrated on EAST with normalized performance projected to achieve steady state operation with 500 MW of fusion power production in CFETR (Q$=$5). In DIII-D experiments, high confinement, internal transport barrier operation that projects nearly to Q$=$10 in ITER at 9 MA has been demonstrated with a fully detached divertor. A synergy between the H-mode pedestal and ITB is found that maintains high global performance as the edge conditions are modified for divertor detachment and heat flux control. Self-consistent simulations predict that, using day-one heating and current drive capabilities, the high-$\beta_{\mathrm{P}}$ scenario in ITER could achieve either mission goals: inductive Q$=$10 performance or steady-state Q$=$5 performance.   

Progress in Disruption Mitigation on the HL-2A tokamak

Mitigation of runaway current was successfully implemented with supersonic molecular beam injection (SMBI) during disruptions deliberately triggered by the massive gas injection (MGI) of argon. A toroidal alfv\'{e}n eignmode (TAE) was observed during disruptions, which plays a favorable role in scattering runaway electrons, and hence, limiting the strength of runaway beam. It has been found that the runaway plateau is easy to obtain on the condition of high normalized magnetic fluctuation level(?B/BT), the runaway plateau is even invisible when ?B/BT the exceeds the threshold of about 7.8\texttimes 10-4, indicating that this magnetic mode plays a scattering role on the RE beam strength. Runaway current caused by argon injection with MGI was successfully suppressed by SMBI with a number of injected helium atoms of about 1.0\texttimes 1021. RE generation during disruptions has been successfully avoided for the first time by the laser blow-off (LBO)-seeded impurity. Metal impurities were injected into the plasma by LBO at 980 ms. With the impurity injection, strong magnetic fluctuation is excited. Plasma disruption was triggered by MGI at 990 ms. It can be observed that no runaway current plateau was formed during disruption. The measurements from a HXR camera show that almost all energetic electrons are lost under strong magnetic fluctuation induced by LBO.   

Simulations and Validation of Disruption Mitigation and Projections to ITER's Disruption Mitigation System

High fidelity 3D initial value simulations of Shattered Pellet Injection (SPI)in DIII-D show that the ablating fragment drives strong parallel flows that transport the impurities along flux tubes and govern the thermal quench evolution. This parallel flow is halted when the ``head bites the tail'', limiting the overall spreading of impurities, and accounting for the observed radiation asymmetry peaking near the injector. DIII-D SPI simulations show that as the thermal quench proceeds, the peak radiation lags behind the ablating fragment and peaks in the accumulated cold impurities that builds up in the wake of the fragment trajectory. Impurity scans of mixed deuterium/neon SPI pellets show a more benign thermal quench due to the enhanced transport and dilution cooling caused by the addition of deuterium suggesting optimal pellet mixtures exist. NIMROD DMS simulations of the Q=10 ITER baseline scenario show that many of the same characteristics are seen in ITER thermal quenches as those observed in DIII-D, particularly the dominance of an n=1 instability in the final thermal collapse. These simulations will be compared along side DIII-D and other tokamak SPI simulations and an initial assessment of the viability of the proposed DMS in ITER will be presented.