26th IAEA FEC Summary : EXC, EXS and PPC Y. Kamada, QST Papers: EXC 101, EXS 57, PPC 26 from 40 devices MEDUSA (Costa Rica ): R~0.14m, a~0.1m, GLAST-III (Pakistan): R=0.2m, a=0.1 m ........ JET (EU): R~3m, a~1.2m W7-X, Welcome to EX sessions ! 2 Welcome to EX ! Start New Operation KTX NSTX-U SST-1 University of Science and Ip~1MA H-modes, Upgraded with Plasma Technology of China H98 ≥ 1, βN ~4 ≥ n=1 no-wall limit Facing Components. RFP with weak/no core MHD R=1.4m, a=0.4m Ip=0.5MA (=>1MA) Bt max=0.35T (=>0.7T) 3 Contents 1. Edge Pedestal System 2. Core Transport 3. Core MHD Stability 4. Operation & Control * Disruption Mitigation is treated in the next talk by Dr. D. Hill 4 H-mode Pedestal Structure and Dynamics Shafranov Shift Fast particles Core Pedestal Plasma Shape (κ, δ, A) ~ Δ B field (3D): RMP plasma response impurity small/no ELM (T, n) (T, regimes Height mitigation Pressure ELM Grassy ELM minor radius Width EDA H-mode EHO, I-mode Turbulence WCM, QCM QH-mode ! Flow Stochasticity Diffusion Heat & Particle Pinch to Divertor grad-p: ELM = Peeling-Ballooning modes (PBMs) width: small-scale turbulence( KBM...) 5 Pressure Gradient ~ Peeling Ballooning Mode Shafranov shift stabilizes COMPASS:The the pedestal gradient, experimental data are in agreement JET and JT-60U: with the EPED confirmed in a wide space model. (EXP6-35, of (κ,δ). Low κ high δ gives lower grad-p, but Komm) COMPASS wider pedestal width, then grassy ELM & good confinement. MHD simulations reproduce (EX/3-4, Urano ) experiment very well. JT-60U and JET: MINERVA-DI: Rotation can destabilize PBMs due to TCV, MAST and TCV minimizing the �∗� effect => JET :The pedestal better fit to exp. data. height has been (TH8-1, Aiba) significantly Multi-machine: JOREK increased by early JET simulations at low increase of βp-core. resistivity/viscosity (EX3-6, Chapman) reproduce experiment (TH8-2, Pamela) 6 ELM crash new key findings KSTAR: Three-stage evolution of ELM Pegasus: J-edge across single was identified using a 2D imaging: (1) ELMs shows the nonlinear quasi-steady filamentary mode with long generation and expulsion of life time n=4-15, (2) abrupt structural current-carrying filaments. transformation into filaments with (EXP4-51, Bongard) irregular poloidal spacing near the onset of crash, (3) and multiple filament bursts during the crash. (EX10-3, Yun) 7 Pedestal evolution during the ELM cycle JET: Pedestal evolution during the ELM cycle: ASDEX-U: 70µs resolution Ti (r) not always consistent with EPED (EX3-3, Maggi) measurement: At ELM, heat flux is Low D2 Gas: first increased at the separatrix, then low-βN: Gradient increases and width constant: Ti(r) becomes flatter. χi comes back not consistent with KBM constraint soon to its pre-ELM neoclassical level . (EXP6-30, Viezzer) Is J-edge expelled by an ELM crash? ? 8 Pedestal Width: key = ne(r) relative to Te,i (r) ASDEX-U: D fueling shifts density profile JET: Pedestal stability improves with outward , and T profile anchored at separatrix reduced radial shift. JET-ILW tends to => causes a significant degradation of the have larger relative shift than JET-C. pedestal top pressure. (EX3-5, Dunne) (EXP6-13, Giroud) Te ped p ne 19 -3 ne,sep(10 m ) pe 9 L-H Transition Threshold Power DIII-D: Dual Mode Nature of JET: Isotope Effect: Non- Pegasus: Ultralow-A Edge Turbulence May linear mass dependence on At low A(~1.2), PLH >> ITPA Explain Isotope and Density L-H power threshold scaling by one order of Scaling of L-H Power (EX5-2,Hillesheim) magnitude.(OV5-4, Fonck) Threshold (EX5-1, Yan) PD, Nunes) Pegasus NSTX MAST Threshold Power KSTAR: PLH increases with (d) (a)(e) (b)(f) (c) δB for any cases with n=1, Ion mode n=2 or mixed-n. (EXP4-4) Ion mode Electron Electron mode Electron Electron Electron mode mode mode mode Ion mode n ~1.5x1019m-3 n ~1.5x10199m-3-3 n ~4.2x1019 9m-3 -3 n ~5x109m-3 e ne~4.2x10e m nee~5x10 m e 10 L-H transition: Behavior of turbulence JET: Radial wavelength of DIII-D: The main-ion poloidal Stationary Zonal Flows scales flow acceleration is quantitatively with the radial correlation consistent with Reynolds-stress- length of turbulence, driven shear flow amplification ~ several times smaller ( EXP3-11, Schmitz) than the width of the edge radial electric field well. (EX5-2, Hillesheim) NSTX: The energy exchange between flows and turbulence was analyzed using GPI .The edge fluctuation do not vary just prior to the H-transition. => Turbulence depletion is probably not the mechanism of the L-H transition in NSTX. (EX5-3, Diallo) ASDEX-U: L-I Transition: Negligible contributions of ZFs. (EXP6-29, Putterich) 11 ELM-free regimes : extended remarkably I-mode QH-mode Alcator C-mod : DIII-D: Discovered Stationary High energy confinement with Temperature pedestal, Quiescent H-mode with Zero Net L-mode Density pedestal, Small impurity, Stationary NBI Torque in double-null shaped and ELM-free. Extended to full field 8T and current plasmas, characterized by 1.7MA. Confirms weak L-I threshold dep. on B, and increased pedestal height & width: wide power range at high B. (EX3-1, Hubbard) sustained for 12τE with excellent confinement (H98y2 ~ 1.5, �N ~ 2). 7.8 T ,1.35 MA (EX3-2, Chen) decreased H =1 98 edge ExB T =7.3keV e0 shear enables destabilization <p>=1.4 atm of broadband turbulence 12 Pedestal fluctuations : variety of interplay EAST: A new stationary small/no KSTAR: Broadband HL-2A: EM turbulence was ELM H-mode was found at low turbulence induced by RMP excited by locally- ��∗ < 0.5, H98 ≳ 1.1, exhibiting a damps the ELM amplitude accumulated impurities. low-n electro-Magnetic Coherent (EXP4-15, Lee) Double critical gradients of Mode. It appears at the low impurity density were frequency boundary of TAE gap. observed and reproduced (+ ELM pacing)(EX10-2, Xu) by theoretical simulation. (OV4-4, Duan) HL-2A: Synchronization of GAMs and magnetic fluctuations was observed in the edge plasmas. (EXP7-27, Yan) 13 Remaining Issue: How is the Pedestal Width determined ? ? (T, n) (T, Pressure ? When pedestal grad-p is below Peeling-Ballooning limit, how does the pedestal width evolve and saturate ? During the ELM cycle? Controllable? 14 Success of RMP ELM Suppression = Phase and Shape DIII-D: ELM control requires KSTAR: Optimal phasing for ASDEX + DIII-D: ELM the applied field to couple to n=1 RMP is consistent with Suppression was obtained an edge stable MHD mode, an ideal plasma response for the first time in AUG at directly observed on high field modeling.(EX1-3, In) low ν* with a plasma shape side. The response is matched to DIII-D (δ~0.3) inversely proportional to ν*. showing the importance of (EX1-2, Paz-Soldan) stable edge kink response. (AUG : EXP6-25, Willensdorfer (PD, Nazikian) MAST: peeling, OV5-3, Kirk ) EAST: n=1 RMP, Plasma response behaves a nonlinear transition from mitigation to suppression of the ELMs (EXP7-4, Sun) 15 Long Pulse ELM Suppression by RMP: Big Progress DIII-D: Fully Noninductive plasmas KSTAR: n=1 RMP ELM suppression was sustained for with high � (≤ 2.8%) and high more than ~90 τE (H89=1.5), and also confirmed to be confinement (H≤ 1.4) sustained for compatible with rotating RMP, wide q95 (4.75 – 5.25) ≤ 2 current relaxation with ECCD (EX1-3, In), (PD, Jeon) and NBCD, and integrated with ELM suppression by n=3 RMP; the EAST: n=1 RMP ELM suppression in long-pulse (> 20s) strong resonant interaction allows was realized with small effect on plasma performance ELM suppression over a wide (H98>1) (P7-4, Sun) range of q95 (EX4-1, Petty) DIII-D KSTAR EAST 16 Core Plasma Transport issues for ITER & DEMO Te/Ti ~ 1 ( <= electron heating (α, high energy NB, IC, ECH), high ne) Electron Transport Small rotation due to small external torque ( => intrinsic torque ) Small central fueling ( high energy NB) => density profile ? Confinement performance with metal divertor can be recovered? Accumulation of heavy impurity (metal wall) ? Isotope Effects on Confinement? 17 Thermal Transport at high Te/Ti DIII-D and JT-60U: Positive Shear (PS) shows JET: High Te/Ti plasmas: Electron reduction in Ti when ECH is added. Negative transport evaluated with linear Central Shear (NCS) minimizes confinement gyro-kinetic simulations GENE: degradation even with increasing Te/Ti~1. DIII- most consistent with D shows smaller rise in low-k turbulent ( ITG/TEM) + ETG. fluctuations in NCS than PS. (EX8-1, Yoshida) => Multi-scale non-linear gyro- DIII-D / JT-60U kinetic simulation underway. (EXP6-14, Mantica) /Ti} /Ti} Te Te T /T ~1 {Low e i {High i i χ χ qegB DIII-D magnec shear Fluctuaon /Ti} /Ti} Te Te PS {Low {High ñ ñ NCS Σ Σ R/LTe 18 Electron thermal Transport NSTX: Electrostatic low-k Gyrokinetic W-7X: Te profile shape follows the ECRH Simulation (GTS) explains ion thermal Power deposition transport , but is not able to explain electron -> no indication of profile stiffness transport. => high-k ETG / EM is important for (EX4-5, Hirsch) electron transport. Nonlinear GYRO simulation explains grad-n stabilization of ETG, but not enough => EM? (EXP4-35, Ren) electron MST: Drift wave turbulence MAST: Fluctuation (TEM) emerges in RFP measured at the top of plasmas when global pedestal is consistent tearing instability is with Electron transport reduced by PPCD. evaluated with linear (EXP5-17, Brower) gyro-kinetic simulations GENE: consistent with ETG.