Alcator C-Mod ICRF Research Program

Transcription

1 Alcator C-Mod ICRF Research Program MIT Plasma Science and Fusion Center February 4-6, 2009 S.J. Wukitch Overall Themes 1. Develop ICRF heating and flow/current drive actuators for optimization i i of fhigh hperformance plasmas. 2. Experimental validation of advanced simulation tools scalable to ITER and reactors. Outline: 1. Context of C-Mod ICRF program 2. Overview of ICRF system and capabilities 3. Proposed research 1. ICRF 2. ICRF-LHRF interactions

2 ICRF Challenges Impacting Utilization ICRF heating has been experimentally demonstrated to be effective and is planned to be utilized in both ITER and future devices. Wave propagation and absorption. Assess ICRF as potential flow/current drive actuator. Physics and simulation validation. Interaction at the plasma edge. Antenna performance. Impurity production. Robust long-distant coupling. Load tolerance. Development and validation of antenna simulation code. Voltage and power handling. Antenna conditioning. Sources are 2 MW and most efficient of all auxiliary heating power sources. Lifetime issues associated with high power tetrode P RF (MW) W MHD (MJ) T e0 (kev) n e (x10 20 m -3 ) 2 R neut (x10 14 s -1 ) P Rad (MW) B T =5.4 T, I P =1 MA Time (s)

3 C-Mod ICRF Program Primary goal of ICRF physics program is to: Provide first principle understanding of ICRF physics including antenna coupling and wave absorption and Develop a reliable heating and current/flow drive actuator that can be utilized to optimize overall plasma performance with minimum negative impact on plasma. ICRF provides bulk auxiliary heating power in C-Mod. Fundamental minority, mode conversion, and second harmonic minority ion cyclotron scenarios are extensively investigated and have begun to investigate Fast Wave. Emphasize validating physics and computational models thru comparison of simulations to experiments. Access to wide range of RF absorption scenarios, diagnostics, and advanced simulation codes. We investigate and develop solutions to technological and physics issues associated with the antenna/coupler and operations to enable successful RF operation.

4 High Priority Research Issues High priority research issues are issues we are well positioned to investigate and C-Mod can make a unique contribution. High leverage physics issues, support for ITER, and student theses. Assessment and develop fundamental understanding of mode conversion flow drive. (Y. Lin) ICRF compatibility with metallic walls (S. Wukitch) Identify primary ICRF impurity source locations. Characterize impact of ICRF power on SOL density profile (C. Lau). Characterize ICRF sheaths with additional emissive and B-dot probes. Rotate antenna to reduce impurity i production. Validation of physics and simulations in concert with RF-SciDAC. Importance of finite banana width on RF absorption (A. Bader) Validation of TORIC in mode conversion with PCI measurements (N. Tsujii) TOPICA validation with loading, antenna impedance, and SOL density profile measurements. (C. Lau)

5 Secondary Research Topics Validation of ITER scenarios, particularly non-activated phase, is under assessment. Hydrogen plasmas are generally more time consuming due to H fraction issues following a particular hydrogen run. He plasmas are more compatible but performance is often worse than in D plasmas. Fast wave heating and current drive will likely have limited run time. Expected decrease in source power while switching over to all 4 strap antennas, although 09 configuration is good. Limited plasma parameter space. Sawtooth stabilization with MCCD and ICCD. Both MCCD and ICCD will likely ramp up as MC flow drive experiments are completed. Antenna power and voltage handling studies are proceeding in test stand (M. Garrett and T. Abram). Have begun investigation of new materials in effort to improve antenna voltage handling.

6 ICRF Antenna Configuration (FY 09) D & E antennas D E F G D & E Antenna Ip GH Full Limiter C LH Coupler H J antenna B AB Split Limiter A J K midplane Limiter K J antenna Frequency 80 MHz MHz Power 2 x 2 MW 4 MW Antenna 2 x 2 Strap 4 Strap Phase fixed variable

7 ICRF Antenna Configuration (FY 10) E antenna E D & E antennas D Ip F G GH Full Limiter New 4-strap antenna installed in J port ( ). This is a rotated antenna designed to lower the impurity production. C LH Coupler B AB Split Slit Limiter A K midplane Limiter K Frequency 80 MHz MHz Power 2 MW 4 MW Antenna 2 Strap 4 Strap Phase fixed variable J H J antenna D antenna removed for diagnostics displaced d from B port. Moves the ICRF farther away from the LH coupler to reduce IC-LHRF edge interaction. Available source power reduced to 6 MW from 8 MW. Lowers coupled power further because ICRF antenna is likely to have lower power handling. Real time matching (double stub FFT system ) on E antenna only.

8 ICRF Antenna Configuration (FY 11) D E E antenna Ip F G GH Full Limiter New 4-strap antenna installed at E port. If rotated antenna is successful in reducing impurities, raising the power density limit will become increasingly importance. C LH Couplers H J antenna Available source power will be increased to 8 MW. B AB Split Limiter A J K midplane Limiter K Frequency 80 MHz MHz Variable phasing available for both antennas. Real time matching will be lost until second unit can be fabricated. Power 4 MW 4 MW Variable frequency for all Antenna 4St Strap 4St Strap transmitters (proposal budget) Phase variable variable

9 Proposed ICRF Antenna Underlying cause of impurity generation is thought to be generation of E. Rotate t antenna structure t 10º to be perpendicular to total B field. Along a field line E will cancel due to symmetry yfor dipole operation and is expected to reduce impurities. Power density at 2 MW (3MW) is 9.8 MW/m 2 (148MW/m ( ) Peak n φ = 14 (0,π,0,π); 11 (0,π,π,0); and 8 (0,π/2,π,3π/2) in vacuum spectrum (bit higher than J antenna). Feedthrus are 5 diameter (present are 4.5 ). Strip line impedance is 30 Ω (J is 50 Ω and D/E is 30 Ω). Screen is aligned to B-field and is 50% transparent (same as J).

10 ICRF Simulation Tools Codes: Microwave Studios and COMSOL finite element electromagnetic commercial codes are available. TOPICA:3D 3-D modeling of ICRF antenna code with full wave plasma model (TORIC) in collaboration with Polytechnico di Torino and RF Sci-DAC (CSWPI). TORIC for wave propagation, power deposition, and current drive calculations.» Coupled with Fokker Planck codes, DKE, and CQL3D. Access to finite banana width Monte Carlo code with self consistent t RF wave fields through h RF SiDAC Sci-DAC. Synthetic Diagnostics: Synthetic phase contrast imaging diagnostic to model [a.u.] measured density fluctuations in TORIC (N. Tsujii). 0 Synthetic charge exchange neutral particle analyzer -1 ( kev) implemented in CQL3D (A. Bader) Plan to implement synthetic charge exchange recombination spectroscopy for fast ions in CQL3D.(thesis project) [a.u u.] 2 1 LCFS IBW ~ Re( ne dl) ~ ne dl 1 Experimentali Synthetic ICW ICW MC layer FW R (m)

11 Diagnostics Present status: 32 channel PCI diagnostic for density fluctuations associated with RF waves. (N. Tsujii) 4-channel CNPA (compact neutral particle analyzer) at F and 8 channel at J. (A. Bader) Rfl Reflectometer t microwave electronics to be delivered 02/09. Plans: SOL reflectometers in new LH launcher and 4-strap antennas (ORNL, C. Lau). Edge probes at RF limiters and plasma limiters: emissive and RF magnetic probes. Implement charge exchange recombination spectroscopy (CXR- FI, equivalent of FIDA on DIII-D) D) for fast ions (UTexas, student thesis). 8 Channel CNPA

12 ICRF Mode Conversion Flow Drive (MCFD) Goals: Characterize mode conversion flow drive. Develop understanding of mode conversion flow drive such that one can reliably predict future experiments and devices. Status: Observed toroidal rotation, Vφ, profile is peaked. ΔV φ is approximately twice the empirical intrinsic rotation scaling And scales as P RF / <n e > not with RF E-field. Change in poloidal rotation profile is peaked off-axis in the ion diamagnetic drift direction. V θ is up to 1.5 km/s and localized: 0.3 < r/a < 0.7. Rotation direction is independent of the antenna phase. Observation contradicts previous theoretical understanding. ICW - ion interaction is key to MCFD. ICW is detected by PCI. To oroidal Rotatio on V φ [km/s] Mode Conversion Minority 80 Ohmic r/a 3.0 -Δ ΔV θ [km/s] Mode Conversion Minority TORIC indicates significant power to the ions when -1.0 rotation tti is observed Experimentally, no flow drive has been observed when r/a majority of ICW power is absorbed by electrons. Ion diamagn netic drift direc ction Y. Lin et al., Phys. Rev. Lett. 101, (2008). Y. Lin et al., post deadline 22 nd IAEA Geneva (2008). Y. Lin et al., APS invited 50 th APS (2008).

13 MCFD: Plans Examine dependence of flow drive on 3 He concentration. Investigate importance of ion versus electron absorption. Investigate flow drive scaling with plasma density. Appears to scale inversely with density. Density also changes the ICW perpendicular wavelength, hence the radial wave pattern. Participate (pending approval) in JET experiments investigating MCFD. Data mining has produced evidence consistent with analysis of JET plasmas. Examine MCFD in H-mode plasmas. Assess influence of plasma current on flow drive. As plasma current decreases, expect more power to IBW than ICW. Intrinsic rotation scales inversely with plasma current. Scan magnetic field around 5 T to investigated dependence on resonance location. Simulation suggests power partition between IBW and ICW varies depending on resonance location. Investigate MCFD in He and H majority plasmas. Examine dependence on wave pattern. ICW perpendicular p wavelength is inversely proportional p to Alfven speed. Compare MCFD at 5 T, 50 MHz and 8 T, 80 MHz. ω meff neff Is flow drive dependent on wave parallel velocity? k ~ B Phase independence suggests that it is not.

14 ICRF Compatibility with Metallic Walls Goals: 150 Identify location of RF impurity sources. H-mode Develop understanding of underlying physics. L-mode H-mode with Examine antenna designs to minimize RF 100 BN tiles sheaths. Assess mitigation techniques. Status: Enhanced erosion/impurity production is 50 localized to the active antenna. P1/2 Characterized ICRF sheath dependence on ICRF power and confinement mode (L,H). Plans: RF Power [MW] Boron coat outer divertor shelf tiles and limiters.» Identify erosion locations. Characterize impact of ICRF power on SOL density profile (C. Lau).» Assess gas puffing effectiveness for modifying i the SOL density profile (IOS-5.2) 52) Characterize ICRF sheaths with additional emissive and B-dot probes.» Re-instrument plasma limiters in first phase.» Additional instrumentation with new 4-strap antenna. l [V] Plas sma Potentia , 22,

15 ICRF Compatibility with Metallic Walls: Plans Compare rotated and standard antenna characteristics. Impurity and gas production. ICRF impact on SOL density profile. sheaths characteristics for range of RF absorption scenarios. Develop analysis tools for investigation of sheath mitigation techniques through collaboration with RF-Sci DAC CSWPI. First step is to utilize lossy dielectric in commercial codes like Microwave Studios and COMSOL. Utilize TOPICA coupled to FELICE (1-D) and cold plasma model in SOL region with real density profiles. Replace cold plasma model of SOL with finite element full wave solver in analysis package. Assess what makes RF sources the dominant core Mo contributor.

16 Wave Propagation: Validate ICRF Simulations Goal: Validate simulations against wide range of experimental regimes. Scenario Characteristics i Status D(H) Mode conversion Strong single pass absorption, Fields toroidally localized Electron heating Multiple scale wavelengths Developed 3-channel CNPA Synthetic CNPA implemented in AORSA-CQL3D Upgraded PCI diagnostic calibration TORIC verified with AORSA D( 3 He) Single pass absorption is ~10% Found L-mode heating effectiveness and H-mode threshold were similar to D(H) heating. Fast twave Electron heating Observed first fast wave heating in C- Single pass from 1-10% Mod 2 nd 2 nd harmonic H at low field Access to upgraded simulation Harmonic Magnetic field scan to investigate capability including finite orbit effects. 2 nd harmonic D absorption.

17 Wave Propagation: Validate ICRF Simulations Plans D(H) 1 st and 2 nd harmonic Measure tail energy and spatial distribution Examine tail formation time (RF-SciDAC) Scan impressed n φ spectrum. Mode conversion Scan minority concentration from minority to mode conversion regime Utilize D( 3 He), D(H), and H( 3 H)di He) discharges. Measure wavenumbers, wave amplitude, wave spatial distribution, and deposition profiles. D( 3 He) Direct comparison of D(H) and D( 3 He) H-mode minority discharges. Alternately move additional resonances in and out of the plasma and compare performance. Experiment TORIC Fast Wave Vary target plasma temperature and electron β. Scan impressed n φ spectrum. nc) S (MW/m^3/MW_i r/a

18 ICRF and LHRF Interactions: Coupling Goal: 0.4 Demonstrate compatibility between ICRF and LHRF to enable tokamak performance 0.3 optimization. Status: 0.2 Coupled LH power into H-mode and L- mode ICRF heated discharges. 0.1 LH faults are significantly increased with neighboring ICRF antenna operation. Gas puffing in coupler box had mixed results Plans: Measure local density profile with reflectometer. Examine influence of ICRF and LH modification of fthe SOLdensity and density profile. Investigate influence of boronization and gas puffing on coupling. Examine density and power dependence of reflected power fraction. Refl. Power Fr raction (Γ 2 ) 0.5 L and H-mode LH Coupling H-mode (LSN) L-Mode (USN) n grill (x10 18 m -3 )

19 IC and LH Waves Interactions: Fast ions and LH Goal: Evaluate fast ion absorption of LH waves (important issue for ITER). Motivation: For effective LH current drive in reactors, parasitic absorption of the LH wave on fusion α-particles needs to be limited. A secondary issue is absorption by fast ions generated by ICRF. JET reported interaction at ¼ the expected energy but latter experiments failed to reproduce results. Tore Supra has not observed interaction. Status: Experiment was complicated by x-ray sensitivity to plasma density. Plans: Key is to identify plasmas where then density is under better control» He discharges may prove to be better targets. Inject fixed LHRF power, scanning n from ( ) on different discharges until an interaction is observed with the minority tail: Measure hard X-ray profile to evaluate the effect on the generation of fast electrons and CNPA to assess impact on fast ion distribution. Model process with GENRAY CQL3D. Scan ICRF power to vary the minority tail energy. Change B to move the ICRF resonance position. Vary the LH wave n to change the LH wave phase speed.

20 Summary Proposed ICRF physics program s goal is to: provide first principal understanding of ICRF antenna coupling and wave absorption physics such that ICRF is a reliable heating and/or current drive actuator with minimum negative impact on the plasma. High priority research issues are: Mode conversion flow drive assessment and characterization, ICRF compatibility with metallic walls, and simulation validation. Second tier research issues have lower priority unless developments warrant more resources. Experiments to validate ITER scenarios, particularly non-activated phase. Evaluate Fast wave heating and current drive for central seed current. Sawtooth stabilization with MCCD and ICCD. Antenna power and voltage handling studies are proceeding in test stand. ICRF-LHRF research issues are: Compatibility of LH and ICRF coupling and Fast ion absorption of LH waves.

21 Reference Material: Summary of Tasks RF Sources: Upgrade ICRF transmitter control and system to improve reliability. Monitor tube lifetime issues. Mthi Matching network: Continue to test and later implement prototype fast ferrite matching network. Antennas: Manufacture and commission two new 4-strap ICRF antenna. Reduce RF sheaths through antenna design. Investigate first fault initiation and HV degradation with B-field. Codes: TOPICA (on MARSHALL) 3-D modeling of ICRF antenna code (U. Torino). TORIC coupled with CQL3D Fokker Planck code for current drive calculation. Finite banana width Monte-Carlo code with self consistent RF wave fields. (Sci- DAC initiative) Diagnostics: Absolute calibration of PCI. SOL reflectometers CNPA CXRS-FI Emissive and RF probes.

22 Reference Material: Context of C-Mod s ICRF Program Standard scenarios are fundamental and second harmonic with access to direct fast wave. High and low single pass absorption scenarios are accessible Antenna power density is ~10 MW/m 2. Utilize ICRF with high Z first wall materials. ITER will utilize fundamental 3 He and second harmonic tritium. ITER is expected to have high single pass absorption. ITER power density is expected to be 6-8 MW/m 2. ITER and future burning plasma devices are likely to utilize high Z first wall materials Proposed research is well aligned with FESAC planning committee recommendations. Make contributions to model validation, Develop and test ICRF antennas, and Evaluate ICRF compatibility with first wall materials.

23 Reference Material: MCFD Scales with RF Power D(3He) MC D(H) Minority [km/sec] 60 [km/sec] 60 ΔVφ ΔVφ ΔW/Ip [kj/ma] P RF [MW]/<n e >[1020 m-3] For D( 3 He) MC absorption, the change in toroidal rotation, ΔV φ, is approximately twice the empirical intrinsic rotation scaling. 1 For D(H) minority heating, the ΔV φ scales with ΔW/I 1 p For D( 3 He) MC absorption, the change in toroidal rotation, ΔV φ, is scales as P RF / <n e > not with RF E-field. 1. J. Rice Nucl. Fus. 41, 277 (2001).

24 Ref. Mat.: MCFD is Independent of Antenna Phase The direction of rotation is independent of the antenna phase. V φ in co-current direction. V θ in ion diamagnetic drift direction. Phase scan showed only 10% variation between cocounter- and heating phase. The rotation magnitudes are similar for all antenna phases.

25 Ref. Mat.: Measurements Confirm Presence of MCICW Mode converted ion cyclotron wave (MCICW) detected by phase contrast imaging about ~ 4 cm away from the 3 He cyclotron resonance and on the HFS of magnetic axis. Wave number k R ~ 3-7 7c cm -1,consistent sste t solution of dispersion equations and to previous MC experiments and also to the [Y. Lin et al, PPCF (2005)].

26 Ref. Mat.: TORIC Indicates Significant Power to Ions MW/m ECE 4 10 Fit 8 3 Simulation r/a Using a n( 3 He)/n e ~ 8-12% for TORIC simulation can reproduces measured MCICW profile from PCI And electron power deposition profile. The MC ICW is damped strongly onto 3 He ions through a substantially broadened IC resonance. Fast wave: k ~ 10 m -1 MCICW: k ~ m -1 0 MW/m 3 ω = ω c + k, 3 v He

27 Ref. Mat.: Power to Ions Appears to be Important Experimentally, no flow drive has been observed when majority of power is absorbed by electrons. [km m/s] 40 2 Flux averaged power deposition to 3 He 20 from TORIC simulation is ΔV θ 1 approximately ate at the same location o 0 0 where: 10 3 MC ICW power to He (flux surface Toroidal rotation profile decreases averaged) significantly and Poloidal rotation is observed and Suggests ICW - ion interaction is key to MCFD. MW/m V φ r/a 4 3 [km m/s] (a) (b)

28 Reference Material: ICRF Current Drive Goals: Tailor current profile with MCCD or ICCD to mitigate or control sawtooth period. id FWCD is being evaluated to provide central seed current for optimizing current profile in AT plasmas. as. Status: Observed sawtooth period control with mode conversion current drive (MCCD) without energetic particles present. Obtained target discharge (T e > 4 kev) to allow evaluation of fast wave (FWCD) and ion cyclotron current drive (ICCD) LHCD ~ ka off AT axis MCCD Sawtooth pacing ST pacing < 50 ka physics FWCD ICCD RF Power [MW] 4 T e0 (kev) 3 2 Central seed current < 50 ka Sawtooth pacing < 50 ka Plasma Pressure [atm] AT H-mode Time (s)

29 Ref. Mat.: Mode Conversion CD Status: Inefficient for bulk current drive. Ehst-Karney parameterization overestimates drive current due to incorrect wave polarization. Good candidate for sawtooth pacing in C-Mod» Clear sawtooth variation with phase and deposition location. Plans: Utilize MCCD to pace sawteeth in the presence of energetic ions driven by ICRF (ITPA MDC-5). Can increasing the local shear pace ST in the presence of energetic ions which increase ST stability? P RF (MW) ctr-cd D-port D+J-port D-port D+J-port 3 T e0 (kev) Time [sec] Time [sec] co-cd

30 Ref. Mat.: FWCD Status: Initial experiments conducted to 4 observe Fast Wave electron 2 heating 0.1 Plans: 0.05 Measure power deposition profile variation with antenna phasing and 4 target discharge temperature and 2 density. 1.5 Compare counter and co- current 1 drive phasing with full compliment of diagnostics. 4 Benchmark TORIC simulation 2 against measured driven current 4 and power deposition profiles. 2 B T ~5.2 T, I p =1.2 MA, USN P RF (MW) FW H minority W MHD (MJ) T e0 (kev) n e (10 20 m -3 ) neutrons (x10 13 s -1 ) P RAD (MW) Time (s)

31 Ref. Mat.:Ion Cyclotron Current Drive Classical regime: Current is carried by passing particles. Peak efficiency near E critical (~15T e for H) Driven current profile is bipolar about the cyclotron resonance. For co-current drive phasing, is stabilizes ST for a cyclotron resonance at the q=1 on the low field side of magnetic axis. Finite orbit (FO) regime: Current is generated by trapped particle orbits. Efficiency increases with particle energy. Driven current profile is bipolar about the cyclotron resonance. For co-current drive phasing, FO-ICCD destabilizes ST for a cyclotron resonance at the q=1 on the low field side of magnetic axis. Access in both L and H-mode discharges. Access in low density L-mode discharges. Plans: In L-Mode low density discharges, access both FO and classical regime. Evaluate ICCD in H-mode discharges in classical l regime and assess efficacy. Use one antenna in CD or CTR-CD phasing with other antenna in heating phase to maintain plasma parameters as similar as possible. Scan BT to sweep cyclotron resonance through q=1 surface on both low and high field side. Monitor changes in ST period and CNPA for tail energy.

32 Ref. M.: IBW and LH Interactions Goal: Utilize ICRF to localize LH wave damping off-axis. Background: IBW waves have been suggested to provide a seed electron population that results in damping of LH waves away from location predicted if electrons have a Maxwellian distribution. Mode conversion heating is at sufficient i high h power density that t the distribution function is likely modified. Can MC heating modify the location at which LH waves are damped? Plans: Utilize MC heating off axis and LHCD at n =3.1 (120 o phase). D+E onaxis minority heating. Measure hard X-ray yprofile to evaluate the effect on the LH current profile and PCI to monitor MC waves. TORIC/CQL3D modeling using self-consistent distribution functions.

33 Ref. Mat.: ICRF Coupling and Active Matching Goals: Maintain maximum power delivered over wide range of plasma conditions. Minimize discharges lost to poor match. Status: Operated up to 1.85 MW with prototype fast (~1 msec) ferrite tuners and routinely operated. load variation associated with ELMs and confinement transitions with reflected power coefficient <3%. Plans: Measure local l density profile and compare with TOPICA as part of ongoing model validation. Optimize tuner characteristics for experimental frequency range. Implement fast ferrite tuners on all antennas H-mode D Antenna (MW) E Antenna (MW) J Antenna (#3) (MW) H-mode Γ 2 =P refl /P forw Γ 2 Γ 2 J Antenna (#4) (MW) Γ n l (x10 20 m -2 ) Time (s) Test methods for edge density and density profile control.» Local gas puffing is an obvious choice for affecting the far scrape-off density.» Edge pedestal modification through magnetic geometry and pumping Time (s)

34 Ref. Mat.: Antenna Power and Voltage Limits Goal: Obtain high power over wide range of plasma conditions. 600 Status: Achieved antenna power densities of ~10 MW/m 2. [MV/ /m] E b ICRF pulse MWf for sec in 200 He, L-mode plasma Discharge formation is responsible for neutral pressure limit. Plans: On test stand, we plan to investigate» Refractory metals breakdown limit, C» Influence B-field on breakdown limit, and» Initial RF trip at neutral pressure limit. Assess new ICRF antenna power and voltage limits. Continue benchmarking of TOPICA. Investigate limitations on 6 MW discharges. A. Descoeudres et al., Proc. EPAC08. Gdcp CuZr Al HC Cu td Cu OFH ht W Ta WC Nb Mo Cr V Ti SS