Comparison of Private Flux Region and High Field Side MGI (massive gas injection) with the conventional LFS MGI in NSTX DRAFT RUN PLAN ITER TSG & MS TSG 8 June, 2011 PPPL, Princeton, NJ NSTX Supported by College W&M Colorado Sch Mines Columbia U Comp-X General Atomics INL Johns Hopkins U LANL LLNL Lodestar MIT Nova Photonics New York U Old Dominion U ORNL PPPL PSI Princeton U Purdue U SNL Think Tank, Inc. UC Davis UC Irvine UCLA UCSD U Colorado U Maryland U Rochester U Washington U Wisconsin Culham Sci Ctr U St. Andrews York U Chubu U Fukui U Hiroshima U Hyogo U Kyoto U Kyushu U Kyushu Tokai U NIFS Niigata U U Tokyo JAEA Hebrew U Ioffe Inst RRC Kurchatov Inst TRINITI KBSI KAIST POSTECH ASIPP ENEA, Frascati CEA, Cadarache IPP, Jülich IPP, Garching ASCR, Czech Rep U Quebec 1 This work is supported by US DOE contract numbers FG03- 96ER5436, DE-FG02-99ER54519 and DE-AC02-09CH11466

NSTX MGI XP Overview of planned experiment 2 Predicting and controlling disruptions is an important and urgent issue for ITER. Reactors based on the ST and Tokamak concepts are expected to carry several MA of plasma current and therefore have the potential for disruption and the generation of substantial amounts of run-away electrons. While work is in progress to avoid disruptions, some may be unavoidable. For these cases, a fast discharge termination method is needed to minimize the deleterious effects of the disruption. Experiments on tokamaks use massive gas injection (MGI) from the low field side to terminate discharges On NSTX, a comparison will be made of massive gas injection from the private flux region and the high field side region to gas injection from the conventional mid-plane injection location for the purpose of safely terminating NSTX discharges. These experiments are expected to add unique new results to the MGI data base.

NSTX MGI XP Theoretical / empirical justification 3 At present Massive Gas Injection (MGI) is the most promising method for safely terminating discharges in ITER. NIMROD simulations for C-MOD and DIII-D have shown that the cold front from the edge, which has been cooled by a massive gas injection pulse, needs to reach the q=2 surface for the onset of rapid core cooling to occur. On ITER, because of the large minor radius of the device, the long transit times for the slow moving neutral gas, and the large scrape-off-layer flows, it is not known if a simple MGI pulse from multiple locations would be adequate. Insight into ways for reducing the total amount of injected gas and optimizing the injection locations would help with the design of a reliable system for ITER. NSTX can offer new data by injecting gas into the private flux and lower x- point regions to determine if this is a more desirable location for massive gas injection.

NSTX MGI XP Theoretical / empirical justification 4 Injection from this new location has two advantages. First, the gas is injected directly into the private flux region, so that it does not need to penetrate the scrape-off-layer. Second, because the injection location is located near the high-field side region, the injected gas should be more rapidly transported to the interior as known from high-field side pellet injection research and from high-field side gas injection on NSTX. By comparing gas injection from this new location to results obtained from injecting a similar amount of gas from the conventional outer mid-plane, NSTX results on massive gas injection can provide additional insight, a new database for improving computational simulations, and additional knowledge to disruption mitigation physics using massive gas injection.

NSTX MGI XP Vary the poloidal location for MGI 5 Initial Experiments (FY11): -Compare MGI into private flux region to mid-plane and to SOL Torr.L gas injection -He + Neon & D2 Detailed Experiments (FY12): -Modify plenum size and valve throughput rates -Consider other poloidal locations -Simultaneous injection from multiple locations to maintain cold edge mantle and reduce poloidal asymmetries 1a: Private flux region 1b: lower SOL 2: Conventional mid-plane injection 3: Variation in poloidal location Unique capability of NSTX: Asses benefits of injection into the private flux region & the high-field side region vs. LFS mid-plane

NSTX MGI XP Experimental Run Plan 6 Part-I: Establish plasma conditions and the trigger times for NBI, MGI & start of VDE Part-II: FY11 experiments Compare PFR vs LFS Compare HFS vs LFS Part-III: FY12 experiments Readjust plenum size & gas throughput based on FY11 results Vary time of gas injection to obtain variation with evolving q surface Install third MGI assembly in position C Simultaneously inject using all three injectors Improve and complete the FY11 data set

NSTX MGI XP Part I: Establish plasma conditions and the trigger times for NBI, MGI & start of VDE 7 Time sequence for initiation of an MGI triggered disruption T0 = MGI TriggerT1 = PF3U/3L Freeze + PF3U Voltage offset T1 = start of VDE Plasma current PF3U PF3L NBI T2 = NBI turn-off time after T2 = T1 & After Ip drops below a certain value T3 = Time when MGI gas front contacts plasma edge Typical estimated Times: T0 = 390ms T1 = 400ms T2 = 401ms T3 = 402ms

NSTX MGI XP Establish VDE and NBI Times – 6 shots 8 Load a 700kA L-mode discharge in Gap Control Test VDE –Freeze the PF3U/L voltage at 400ms (time = T1) –At 401ms impose a +30V increment to PF3U Repeat a total of 3 times –Measure time it takes for plasma to disrupt Adjust T2 (based on Ip value & based on T1) to turn off NBI –Turn NBI off at T1 (T2 = T1, set manually) –Also impose an Ip interlock to turn-off NBI when Ip drops below 650kA. Interlock to take effect 100ms after current flat-top starts to allow NBI to be injected during current ramp Or inject NBI after current flat-top is attained ? Note: This could be done on Day-1 (Piggy back operation during XMP64 or 71)

NSTX MGI XP Establish MGI Time 9 Load a 600kA Ohmic plasma (No NBI) After Ip reaches flat top trigger the Lower Dome MGI –Use Neon at 25, 50 and 100% plenum pressure –Use H α, SXR and bolometers to determine time delay from MGI trigger to evidence for gas contacting the plasma edge Repeat by triggering the Mid-plane MGI Repeat by triggering the slower MGI but with 25% Neon and 75% deuterium (on a different day) Note: This could be done on Day-1, but with NBI turned OFF before MGI injection (Piggy back operation during XMP64 or 71)

NSTX MGI XP Establish Thomson Time for Gas Assimilation studies 10 3 development shots needed to adjust Thomson timing Trigger the two Thomson lasers separated in time by 0.1, 0.25, 0.35 and 0.5 ms to determine the appropriate Thomson timing for gas assimilation studies Need reliable Thomson profiles with good equilibrium reconstructions, but with as much time delay as possible between the two lasers

NSTX MGI XP Part II & III Run Plan 11 For all cases the primary objective is to obtain data for lower dome injection and mid-plane injection using the same amount of gas and by keeping all other conditions identical. Use HEGDC between shots (no Lithium) The time of gas injection will be varied as the q-profile is evolving and this would provide physics information on the importance of the time and spatial dependence of the q=2 surface for initiation of the thermal quench. The toroidal field could also be varied to alter the q evolution. The comparisons that are being made are: -Private flux region injection vs. LFS mid-plane -High field SOL vs. LFS SOL Combination of gas mixtures will be used -Based on Part-I results, use pure Neon or 25% Neon and 75% D2 -Initial results will be obtained in NBI heated L-mode, later discharges in H-mode

NSTX MGI XP PART- IIA1: L-Mode Discharges (total 10 shots) Load L mode shot that has Private Flux Region over organ pipe –Obtain reference shot (1-shot) Neon injection (30, 60 and 100% of plenum pressure) –Then 25% Ne, 75% D2 at 100% plenum pressure –First from LFS injector and then from Lower dome injector (8 shots) –(Order Neon/D2 mixed bottle) –Obtain reference shot (1-shot) 12

NSTX MGI XP PART- IIA2: L-Mode Discharges (total 10 shots) Load L mode shot that has SOL Region over organ pipe –Obtain reference shot (1-shot) Neon injection (30, 60 and 100% of plenum pressure) –Then 25% Ne, 75% D2 at 100% plenum pressure –First from LFS injector and then from Lower dome injector (8 shots) –Obtain reference shot (1-shot) 13

NSTX MGI XP PART- IIB: H-Mode Discharges (total 10 shots) Load H mode shot that has Private Flux Region over organ pipe –Obtain reference shot (1-shot) Based on Part A use pure Neon or Neon + D2 –First inject from the mid-plane location, then from the lower divertor region (4 shots) –Obtain reference shot (1-shot) Load H mode shot that has SOL Region over organ pipe –Obtain reference shot (1-shot) Based on Part A use pure Neon or Neon + D2 (4 shots) –First inject from the mid-plane location, then from the lower divertor region (4 shots) –Obtain reference shot (1 shot) 14

NSTX MGI XP PART- IIIA: Private Flux Region vs. Mid-plane (L-mode) q – profile variation (7 shots) Load reference L-mode shot Use 3 different times for gas injection at different values of q –Determine appropriate times from the reference shot –Use Neon or Neon+D2 based on Part-II results Minimum 1 shot at each condition Inject identical amounts of gas at same time from the Lower Dome and Mid-plane injectors for the comparison 2 shots at each time x 3 time values = 6 shots 15

NSTX MGI XP PART- IIIB: Private Flux Region vs. Mid-plane (L-mode) Simultaneous injection from 3 locations (4 shots) Load reference L-mode shot Trigger all three injectors at the appropriate times (based in Part-I results) –Use Neon or Neon+D2 based on Part-II results Minimum 2 shots at each condition Repeat reference shot 16

NSTX MGI XP Run Scheduling 17 First set of experiments (Part-I) to be scheduled early in the run during NSTX XMP-64 or 71 Part-IIA experiments to be scheduled as early as possible in FY11 After that the experiment would benefit from short time slots as they become available

NSTX MGI XP Required machine, NBI, RF, CHI and diagnostics capabilities LSN Plasma in gap control with private flux region above organ pipe location –2 to 3 NBI sources, no RF, no CHI High triangularity plasma with lower SOL region over organ pipe. –2 to 3 NBI sources, no RF, no CHI Key diagnostics –Core diagnostics Thomson scattering, Soft X-ray, Bolometer, Magnetics –Edge diagnostics Hα, Spectroscopic diagnostics –Divertor diagnostics Halo current monitors, Langmuir probes Eroding thermocouples, Two color divertor IR camera Lower divertor visible camera in X-point mode with outer mid-plane views 18

NSTX MGI XP Analysis Reduction in heat loads on divertor (convective heat loads) Reduction of electromagnetic forces associated with poloidal halo currents Anything else? Provide data to groups involved in NIMROD, KPRAD, EIRENE-SOLTPS PPPL computational studies? 19

NSTX MGI XP Planned Analysis Integration of Diagnostics and Resulting data 20 Thomson scattering, EFIT, neutral pressure gauges Physics of gas penetration (fraction that penetrates separatrix) H-alpha array, neutral pressure gauges System response time (gas trigger time to first detection of injected gas interacting with the plasma edge) Multi-color Soft X-ray, H-alpha, Ip, EFIT, Thomson scattering, Mirnov coils Delay in current quench after the gas contacts plasma edge Rate of current quench and vertical dynamics of the plasma 3-D MHD response to the whole equilibrium and MHD activity Thermal quench evolution & pedestal collapse Bolometer array- Core radiated power dynamics Halo current sensors- Dependence on halo current amplitude on gas assimilation (Mitigated vs. beta limit and a VDE disruption) Two color divertor fast infrared camera and Eroding thermocouples Spatial distribution of Thermal loads & fast heat flux measurements Locked mode, RWM mode - n=0 mode detectors - Precursors to disruption