Locked-mode avoidance and recovery without external momentum input using Ion Cyclotron Resonance Heating

Transcription

1 1 EX/P4-39 Locked-mode avoidance and recovery without external momentum input using Ion Cyclotron Resonance Heating L. F. Delgado-Aparicio 1, J. E. Rice 2, E. Edlund 2, I. Cziegler 3, L. Sugiyama 4, D. A. Gates 1, J. Terry 2, S. Wolfe 2, C. Gao 2, T. Golfinopoulos 2, J. Irby 2, R. Granetz 2, Y. Lin 2, S. Wukitch 2, M. Greenwald 2, A. Hubbard 2, J. W. Hughes 2, M. Porkolab 2, E. Marmar 2, S. Houshmandyar 5, P. Phillips 5 and W. Rowan 5 1 Princeton Plasma Physics Laboratory, Princeton, NJ, 08540, USA 2 MIT - Plasma Science and Fusion Center, Cambridge, MA, 02139, USA 3 University of California - San Diego, San Diego, CA, 92093, USA 4 MIT - Laboratory of Nuclear Science, Cambridge, MA, 02139, USA 5 The University of Texas at Austin, TX, 78712, USA Corresponding Author: ldelgado@pppl.gov Abstract: New observations of the formation and dynamics of error-field-induced locked-modes at ITER toroidal fields, without fueling and external momentum input have recently been carried out on Alcator C-Mod. Delay of the mode onset and recovery from pre-existing lockedmodes has been successfully obtained using Ion Cyclotron Resonance Heating (ICRH). The use of external heating concomitant with the n = 1 error-field ramp-up resulted in a delay of the mode-onset avoiding the density pump-out and achieving high-confinement H-modes. Heating the low-density plasma after the mode-onset was not conducive to an L H transition but resulted in unlocking the plasma without external torque and obtaining co/counter-current flows at the edge/core. This simple heating technique could provide an important actuator to circumvent error-field-induced locked-mode disruptions in tokamak plasmas. 1 Introduction The objective of tokamak research is to demonstrate the scientific and technological feasibility of fusion power for world energy production. One fundamental advantage of the tokamak concept is its toroidal symmetry. However, magnetic field perturbations arise inevitably because of departures from axisymmetry due to imperfections or misalignment of the poloidal and toroidal field coils, current feeds to these coils, eddy currents and ferritic material in the vicinity of the plasma. These small deviations from toroidal axisymmetry are well known to destabilize non-rotating tearing modes (also known as locked-modes), which can significantly impact plasma operation. Experimentally, it has been confirmed

2 EX/P that a resonant field component with (2, 1) poloidal and toroidal harmonics can induce a locked magnetic island in both conventional and spherical tokamaks [1]-[8]. Controlled experiments with error-field-induced locked-modes are observed to result in a strong density pump-out due to the effect of the resonant magnetic perturbation, partial or complete braking of toroidal rotation, modification of pressure-driven sawteeth instabilities and a significant reduction in energy and particle confinement, often leading to disruptions and associated vertical displacements. The deleterious effects from these 3D perturbations are more easily produced in low-density plasmas, and so are of most concern for ITER, especially during the early heating phase proposed for high-confinement H-mode access. This restriction has placed design and operational constraints using error-field-correction by 3D coils and various forms of driving plasma rotation for mode stabilization. 2 Physics motivation Experimental locked-mode threshold studies have considered only engineering/global macroscopic parameters resulting in a scaling law of the form Br lock /B T n αn e B α B T qαq 95 R α R 0 [1]-[8]. The determination of this dependence is useful for extrapolating low-aspect and standardaspect ratio tokamak results to ITER. However, the influence of drift-mhd as well as collisional and neoclassical flow-damping effects dependent on local kinetic profiles [9]-[12] can alter the predicted scaling. Including the effects of toroidal rotation (ω φ ), the lockedmode threshold can be modified to δb (2,1) (ω φ )/B T = ( δb (2,1) /B T ) (0.2ωφ /ω i,d ) 3/2, where ω i,d is the ion diamagnetic frequency [12]. This scaling suggests that there is a strong dependence on rotation frequency, offering a window of opportunity for mode stabilization regardless of the torque source (e.g. extrinsic or intrinsic). An obvious actuator for raising the error field threshold is thus spinning the plasma using the external torque imparted by a tangential neural-beam-injection (NBI) system. An increase of up to a factor of two in threshold with NBI has been demonstrated experimentally on DIII-D in L-mode [13], TEXTOR [14] and JET experiments [6, 15]. Unfortunately, enough toroidal momentum density might not be available from the beams to stabilize the mode in ITER. Previous tests on COMPASS-D [6, 16] and new experiments at DIII-D [17] have shown that local Electron Cyclotron Resonance Heating (ECRH) and Current Drive (ECCD) can also stabilize and even remove locked-modes. One small caveat is that ECRH or ECCD stabilization becomes impossible if the mode locks to the error-field or to the vacuum vessel wall in a position not accessible to the EC launcher. Recent experiments in DIII-D used n = 1 magnetic perturbations to control the rotation and toroidal phase of locked modes, positioning the mode in front of the launchers and thereby enabling their suppression. However, after ECCD is turned off, 2/1 locked-modes grow again. A dedicated system of n = 1 fields for locked-mode suppression in ITER may not be feasible since RMP fields will be used for a wide variety of other applications such as error field correction, ELM pacing and suppression as well as RWM stabilization. Therefore, actuators based on global RF heating and current drive - which could also have an indirect effect modifying the underlying momentum transport and toroidal and poloidal rotation - must also be considered for a simpler stabilization approach. The

3 3 EX/P4-39 n e [ m -3 ], T e [kev], I CC [ka], W p [kj] Core V t,ar [km/s] C-Mod # CORE 0.8 a) 1.4 I cc/4 c) 0.6 n e,lo4 /2 τ LM T e,0 / b) τ LM Cocurrent (+) Countercurrent (-) Time (s) W p /100 Ly α -line (H-like Ar) w-line (He-like Ar) n e [ m -3 ] T e [kev] t=0.444 s t=0.494 s t=0.711 s t=1.261 s t=0.444 s t=0.494 s t=0.711 s t=1.261 s d) C-Mod # q=2 r/a~ Radius (m) CORE 2 1 e) T e [kev] (FRC-ECE) C-Mod # q= Radius (m) FIG. 1: (Color online) a)-b) Main plasma parameters from error-field-induced lockedmode discharge in C-Mod. The n e and T e profiles during mode-locking are shown in c) and d), respectively. e) The location of the 2/1 surface has been deduced from the flattening of the T e profile measured by the high-resolution electron cyclotron emission diagnostic. C-Mod s Ion Cyclotron Resonance Heating (ICRH) system provides a unique contribution to worldwide research in support of ITER, providing global heating in ITER-like experiments without NBI particle fueling and momentum input. 3 Experiments in C-Mod Error-field-induced locked-modes can be studied in C-Mod at ITER toroidal fields and without NBI fueling and external momentum input. The typical plasma discharge parameters used were I p 0.8 MA, B φ,0 5.4 T, q 95 4, with central electron temperature and densities of 2.0 kev and ( ) m 3 (see Fig. 3). The safety factor on axis was q and as a result sawteeth activity was present; scenarios with other MHD activity like long- [18, 19] and short-lived [20] modes were not considered. A small nonperturbative concentration of argon was injected to asses the toroidal rotation and ion temperature with an x-ray crystal spectrometer [21, 22] without altering plasma fueling and momentum input. Locked-mode excitation is achieved by ramping-up a set of external control A-coils [see red trace in Fig. -a)] capable of producing non-axisymmetric, predominantly n = 1, fields with different toroidal phase and a range of poloidal mode, m, spectra [7]. The time, τ LM, which marks the change in sawteeth amplitude and frequency as well as the locking-phase of the core plasma were reproducible to within ±65 ms and showed to be sensitive to the density evolution ( n e m 3 ) in accordance with a strong linear dependence of mode-locking thresholds (α n 1). Additional features at τ LM include significant braking of the core toroidal rotation [see Fig. 3-b)], a strong density

4 EX/P pump out due to interaction between the plasma and the resonant magnetic perturbation at nearly the same temperature [see Figs. 3-c) and -d)], and a flattening of the temperature profiles which is measured using the ECE diagnostic at the q = 2 rational surface [see Fig. 3-e)]; the saturated island is approximately 6% of the minor radius. The density pump-out can also be the root for a reduction in the mode-locking thresholds, and is the main cause for a strong reduction in stored energy, confinement time and neutron production [23]-[25]. As a result of the strong pumpout there is also a concomitant decrease in the density fluctuations measured by reflectometers (not shown here) and the ubiquitous appearance, before τ LM, of high- m coherent edge magnetic fluctuations as shown in Fig. 2; τ 0 and τ 1 correspond to the rise-time of the control coils to their maxi- Frequency (khz) τ 0 τ 1 τ LM C-Mod # , BPO_GHK Time (s) FIG. 2: khz magnetic signatures measured at the vacuum wall appear during error field penetration and last for the entire locked-mode. mum current of I CC = 3.5 ka as shown in Fig. 3-a). Although the mode begins with a frequency of khz, it quickly branches out to frequencies up to 70 khz lasting for the entire locked-mode period. These fluctuations are not core-localized since they cannot be observed by high-resolution SXR, ECE or TCI diagnostics but are detected at the outer plasma using magnetic probes and He gas-puff-imaging (GPI) at the edge. EDGE poloidal & toroidal velocity (km/s) a) Before mode-locking v φ v θ b) During mode-locking v θ v φ Toroidal CXRS Poloidal CXRS ρ FIG. 3: CXRS edge poloidal and toroidal velocities a) before and b) during mode-locking; green arrows indicate tendency to lock in time. Shot: Time: s s Shot: Time: s s The toroidal and poloidal flow velocities in the boundary (ρ > 0.9) before and during modelocking transition (shown in Fig. 3) were measured using a transient D 2 gas puff. These CXRS measurements show a decrease from +20 km/s (co-current) to zero over the course of about 70 ms. This is approximately equivalent to the time it takes for the density to pump out as the plasma locks. Measurements of the toroidal velocity at the magnetic axis (ρ 0) - using the high-resolution argon x-ray imaging spectrometer - indicates a much slower but similar slowing-down trend reducing the core velocity from 20 km/s (counter-current) to zero [see velocity time histories inferred from He- and H-like argon emission lines shown Fig.

5 EX/P b) PICRH 2 MW c) PICRH 3 MW Time (s) e) PICRH 2 MW f ) PICRH 3 MW C-Mod # , BPO_GHK 60 C-Mod # , BPO_GHK d) PICRH 1 MW C-Mod # , BPO_GHK Frequency (khz) a) PICRH 1 MW C-Mod # , BPO_GHK Frequency (khz) 60 C-Mod # , BPO_GHK 40 ICRH LM-recovery C-Mod # , BPO_GHK Frequency (khz) ICRH LM-avoidance 1.6 Time (s) FIG. 4: (Color online) Magnetic signatures during ICRH power-scan aimed at a)-c) delaying LM-onset and d)-f ) attempting LM recovery. 3-b)]. After several confinement times the edge and core plasma are assumed therefore to be at rest. The first experiments aiming at delaying the mode onset using ion cyclotron resonance heating (ICRH) are shown in Figs. 5-a), -b) and -c); for these cases heating was applied in-sync with the current ramp-up of the error-field control coils. The use of 1 MW delayed the mode-onset but was not sufficient to transition into an H-mode even when the stored energy was doubled from 25 to 55 kj. However, for the 2 and 3 MW cases the plasma experienced L H transitions as the average density, core ion temperature, toroidal velocity and stored energy increased up to m 3, kev, km/s in the co-current direction (following the Rice s scaling [26]-[28]) and 150 kj, respectively. Nonetheless, when ICRH power is turned off the core plasma locks at later times and its characteristic high-frequency magnetic signatures measured at the wall reappear until the end of the discharge [see Figs. 4-a), -b) and -c)]. The locking times ( ) after the heating pulse are similar for all these cases due to nearly identical time-histories of density, temperature and toroidal flow velocity. H-mode access in the presence of an error-field and ICRF heating appears therefore not be a challenge at the high densities after the plasma Ip ramp-up.

6 EX/P A different set of experiments using transient ICRH power pulses onto the low density, non-rotating phase of the locked-mode aimed at restoring the degraded plasma profiles and its gradients, as well as attempting to unlock the core plasma. These L-mode plasmas did not experience transitions to H-mode due to their lower-density ( 1/2 of the core density before locked-mode onset), even though P ICRH was raised up to 3 MW. During these recovery experiments, the core electron and ion temperature increased by 400, 600, 1000 ev when heated with P ICRH =1, 2 and 3 MW, respectively. The density augments were small and the stored energy increased by 20, 35, 55 kj, respectively. The high-frequency magnetic signatures remained when 1 MW was applied [see Fig. 4-d)] but were suppressed for the cases with P ICRH =2 and 3 MW [compare with Figs. 4-e) and -f)]. A noticeable change in the electron temperature profile before and after the ICRH pulse was recorded using the ECE diagnostic (not shown here). The electron temperature around the 2/1 surface was raised 400 ev, not only restoring but increasing the temperature gradients. Similar trends have been observed heating the electron channel with Lower Hybrid Current Drive (LHCD) and will be subject of a future contribution During the ICRH recovery phase a clear accelera- 0.5 MW tion in the ion-diamagneticdrift direction was observed MW at the edge (ρ 1) with the GPI diagnostic (not shown 500 here). Although the temporally Fourier-resolved signal, S(f), shows an increase after 2.0 MW RF hits the plasma, we find -12 that the conditional spectrum s(k f) does not show MW evidence of a feature in the EDDD propagating at the critical 7 mm depth in ei- V,0 - Z-line [km/s] W MHD /I p [J/A] ICRH heating Counter-current FIG. 5: Changes of the core toroidal velocity and ion temperature during the ICRH locked-mode recovery experiments. Power scan was in the range from 0.5 to 3 MW. T Ar,0 - Z-line [ev] ther of these plasmas. Instead, the growth of fluctuation power in the high frequency components with RF is due to an acceleration in the IDDD (not shown here). Furthermore, this acceleration increases with the amount of applied heating power. The CXRS measurements at the edge (0.9 < ρ < 1.0) also indicate that the edge plasma spins rapidly in the (+) co-current direction up to about 20 km/sec, much faster than the braking phase and recovering the toroidal velocity before the mode onset. Strong changes in the radial electric field, E B drifts and flow velocities are commonly observed when using ICRH [29]. The observed velocities are consistent with the presence of changes in the electric potential arising as a consequence of sheath rectification of the parallel component of the launched waves. A comparison of the change in core (ρ 0) toroidal flow velocity and ion temperature between an Ohmic and

7 7 EX/P4-39 ICRH-heated locked-mode measured with the high-resolution x-ray crystal spectrometer is shown in Fig. 5. One interesting detail is that for every increase in the ion temperature due to heating, the change in core toroidal rotation points always in the ( ) countercurrent direction, unlocking the plasma and recovering the direction and magnitude of the toroidal flow before the formation of the locked-mode [see early phase in Fig. 1-b)], but opposite to the Rice-scaling in the (+) co-current direction. Exploring the connection between typical gradients ( T e,i ) and toroidal rotation (v φ ) through the residual stress (e.g. by changing the underlying turbulence from ITG to TEM affecting anomalous momentum transport [26, 27, 28, 30, 31]) and the effects of neoclassical toroidal viscosity (NTV) can also provide a useful tool to indirectly unlock the edge and core plasma. As mentioned above, the generation of poloidal or toroidal flow velocities is a very attractive parameter for scaling the locked-mode thresholds by its direct relation with the viscous torque. This simple heating technique could provide an important actuator to avoid or circumvent error-field-induced locked-mode disruptions in tokamak plasmas. 4 Conclusions In summary, error-field-induced locked-modes at ITER toroidal fields have been studied at C-Mod without the influence of external fueling and momentum input. Delay of the locked-mode onset and recovery from pre-existing locked-modes has been successfully obtained using ICRH. The use of external heating in-sync with the error-field ramp-up resulted in delay of the mode-onset. Once P ICRH is turned off, the core plasma locks at later times depending on the density and toroidal velocity evolution. In the presence of an error field, an L-mode discharge can transition into H-mode after the current ramp up and still at high densities. For the mitigation experiments, applying ICRF heating to low-density plasmas which have already locked causes the edge/core plasma to spin in the co/counter-current direction, recovering the rotation direction and magnitude that was present before the mode onset, conserving momentum density in the absence of external NBI torques. This work was performed under US DoE contracts including DE-FC02-99ER54512 and others at MIT and DE-AC02-09CH11466 at PPPL. References [1] T. C. Hender, et al., Nucl. Fusion, 32, 2091, (1992). [2] T. C. Hender, et al., Nucl. Fusion, 47, S128, (2007). [3] R. J. La Haye, et al., Phys. Fluids, B4, 2098, (1992). [4] T. H. Jensen, Phys. Fluids, B3, 1650, (1991). [5] R. Fitzpatrick, Nucl. Fusion, 33, 1049, (1993). [6] R. J. Buttery, et al., Nucl. Fusion, 39, 1827, (1999).

8 EX/P [7] S. M. Wolfe, et al., Phys. Plasmas, 12, , (2005). [8] J. E. Menard, et al., Nucl. Fusion, 50, , (2010). [9] A. B. Mikhailovskii, et al., Plasma Phys. Rep., 21, 789, (1995). [10] A. Cole, et al., Phys. Plasmas, 13, , (2006). [11] A. Cole, et al., Phys. Rev. Letters, 99, , (2007). [12] J-K. Park, et al., Nucl. Fusion, 52, , (2012). [13] R. J. La Haye, Rep. GA-A22468, General Atomics, San Diego, CA (1997). [14] P. C. De Vries, et al., Plasma Phys. Control. Fusion, 38, 467, (1996). [15] A. Santagiustina, et al., Proceedings of the 22nd European Physical Society, Bournemouth, Geneva, (1995). [16] P. G. Carolan, et al., Proceedings of the 41st European Physical Society, Montpellier, (1994). [17] F. Volpe, et al., Phys. Rev. Lett., 115, , (2015). [18] L. Delgado-Aparicio, et al., Phys. Rev. Letters, 110, , (2013). [19] L. Delgado-Aparicio, et al., Nucl. Fusion, 53, , (2013). [20] L. Delgado-Aparicio, et al., Phys. Plasmas, 22, , (2015) [21] M. L. Reinke, et al., Rev. Scientific Instrum., 83, , (2012). [22] L. Delgado-Aparicio, et al., Plasma Phys. Control. Fusion, 55, , (2013). [23] L. Delgado-Aparicio, et al., Proceedings of the 41st European Physical Society Conference on Plasma Physics, Berlin, Germany, (2014). [24] L. Delgado-Aparicio, et al., Proceedings of the 56th Annual Meeting of the APS Division of Plasma Physics, New Orleans, Louisiana, (2014). [25] L. Delgado-Aparicio, et al., Proceedings of the 57th Annual Meeting of the APS Division of Plasma Physics, Savannah, Georgia, (2015). [26] J. E. Rice, et al., Phys. Plasmas, 11, 2427 (2004). [27] J. E. Rice, et al., Nucl. Fusion, 44, 379 (2004). [28] J. E. Rice, et al., Nucl. Fusion, 47, 1618, (2007). [29] I. Cziegler, et al., Plasma Phys. Control. Fusion, 54, , (2012). [30] J. E. Rice, et al., Phys. Plasmas, 19, , (2012). [31] R. McDermott, et al., Nucl. Fusion, 54, , (2014).