1 EBW & HHFW Research - G. Taylor PAC-19 2/23/06 EBW & HHFW Research (Including EBW Collaborations with MAST & P EGASUS ) Gary Taylor presented on behalf of the NSTX Research Team NSTX Program Advisory Committee Meeting (PAC-19) February 23, 2006

2 EBW & HHFW Research - G. Taylor PAC-19 2/23/06 Outline NSTX EBW Research (including MAST & P EGASUS ) NSTX EBW Research (including MAST & P EGASUS ) NSTX HHFW Research NSTX HHFW Research Summary Summary

3 EBW & HHFW Research - G. Taylor PAC-19 2/23/06 NSTX EBW Research (including MAST & P EGASUS ) EBW research is directed towards… EBW research is directed towards…

4EBW & HHFW Research - G. Taylor PAC-19 2/23/06 Deposition similar for 14 GHz & 28 GHz and  ~ 20-40% CQL3D/GENRAY NSTX B t (0) = 3.5 kG Need ~100 kA RFCD between  = 0.4 &  = 0.8; Need ~100 kA RFCD between  = 0.4 &  = 0.8; CD localization is not critical CD localization is not critical Current Density (A/cm 2 ) EBWCD Needed to Sustain  > 20 % Non-Inductive NSTX Plasmas Need resilient coupling Need resilient coupling to EBWs for viable EBWCD scheme to EBWs for viable EBWCD scheme Large trapped particle Large trapped particle fraction on low field side fraction on low field side enables off-axis Ohkawa EBWCD enables off-axis Ohkawa EBWCD G. Taylor et al., Phys. Plasmas 11, 4733 (2004)

5EBW & HHFW Research - G. Taylor PAC-19 2/23/06 Strong Diffusion Near Trapped-Passing Boundary Enables Efficient Ohkawa EBWCD Contours of Quasilinear RF Velocity Space Diffusion Operator GENRAY/CQL3D

6EBW & HHFW Research - G. Taylor PAC-19 2/23/06 Synergy with Bootstrap Current Modifies EBWCD Current Profile & Enhances CD Efficiency by ~ 10% R.W. Harvey & G. Taylor, Phys. Plasmas 12, (2005) NSTX  = 40%P EBW = 1 MW, Co-EBWCD NSTX  = 40%, P EBW = 1 MW, Co-EBWCD GENRAY/CQL3D Synergistic bootstrap current results from EBW-induced pitch angle scattering into trapped electron population Synergistic bootstrap current results from EBW-induced pitch angle scattering into trapped electron population Radial diffusion broadening is marginally important for  = 40% case:  ~  ~ 0.7Radial diffusion broadening is marginally important for  = 40% case:  ~  ~ 0.7 CD modeling assuming diffusion constant in velocity space, being performed for range of NSTX conditionsCD modeling assuming diffusion constant in velocity space, being performed for range of NSTX conditions CQL3D being upgraded to improve radial transport calculation, including possible self-consistent anti-pinch effect of Ohkawa EBWCD & its effect on bootstrap current CQL3D being upgraded to improve radial transport calculation, including possible self-consistent anti-pinch effect of Ohkawa EBWCD & its effect on bootstrap current

7EBW & HHFW Research - G. Taylor PAC-19 2/23/06 Measured 80% B-X-O Coupling in L-Mode Edge Plasmas, Consistent with Modeling, Phys. Plasmas 12, (2005) G. Taylor et al., Phys. Plasmas 12, (2005) J. Preinhaelter et al., AIP Proc. 787, 349 (2005) 3-D ray tracing & full wave 3-D ray tracing & full wave EBW mode conversion model EBW mode conversion model using EFIT magnetic equilibrium using EFIT magnetic equilibrium & Thomson scattering T e & n e & Thomson scattering T e & n e f =16.5 GHz NSTX Shot Frequency = 16.5 GHz, B t (0) = 4.5 kG

8EBW & HHFW Research - G. Taylor PAC-19 2/23/06 ~ 30% B-X-O Coupling in H-Modes Consistent with Collisional 3f ce EBE Damping at Upper Hybrid Layer T rad ~ T e (R=1m) Measured T rad Simulated T rad with Z eff = 3 at UHR Simulated T rad with Z eff = 5 at UHR Simulated T rad with no EBW damping at UHR T rad (keV ) 1000 Time (s ) Shot = f = 25 GHz T e ~ 20 eV at UHR, near foot of H-mode pedestal T e ~ 20 eV at UHR, near foot of H-mode pedestal EBW collisional damping sensitive to Z eff for T e < 30 eV EBW collisional damping sensitive to Z eff for T e < 30 eV Measured T rad behavior consistent with EBE only from 3f ce off-axis Measured T rad behavior consistent with EBE only from 3f ce off-axis Li conditioning may significantly improve B-X-O coupling from H-modes by increasing T e at UHR Li conditioning may significantly improve B-X-O coupling from H-modes by increasing T e at UHR T rad ~ T e (R=1.3 m)

9EBW & HHFW Research - G. Taylor PAC-19 2/23/06 M. Carter ORNL AORSA-1D NSTX  = 40% B t (0) = 3.5 kG f = 28 GHz Remotely-steered B-X-O Antennas, Covering 8-40 GHz, will Test EBW Coupling Efficiency Predictions from Models Experiments will study L- & H-mode plasmas Experiments will study L- & H-mode plasmas and effect of Li pumping and effect of Li pumping on EBW edge coupling on EBW edge coupling Scanning frequency & antenna viewing angle Scanning frequency & antenna viewing angle provides key information on angle of peak provides key information on angle of peak EBW coupling & emission polarization EBW coupling & emission polarization EBW coupling study critical for assessing EBW coupling study critical for assessing technical feasibility of EBWCD in FY06 technical feasibility of EBWCD in FY06 Supports EBW T e (R,t) diagnostic development Supports EBW T e (R,t) diagnostic development n pol n tor

10EBW & HHFW Research - G. Taylor PAC-19 2/23/06 MAST EBW Collaboration EBWCD benchmarking between GENRAY/CQL3D & BANDIT EBWCD benchmarking between GENRAY/CQL3D & BANDIT shows reasonably good agreement for  = 40% NSTX case: shows reasonably good agreement for  = 40% NSTX case: - both predict Ohkawa EBWCD, peaked at r/a ~ BANDIT: 26 kA/MW, GENRAY/CQL3D: 37 kA/MW - BANDIT includes EBW coupling at UHR, GENRAY did not - In 2006, will include EBW coupling in GENRAY 28 GHz EBW I p initiation/ramp-up system will be tested on MAST in late 2006 (installation in summer 2006): 28 GHz EBW I p initiation/ramp-up system will be tested on MAST in late 2006 (installation in summer 2006): - collaborate with 28 GHz startup/ramp-up experiments in collaborate with 28 GHz startup/ramp-up experiments in possible 28 GHz B-X-O experiments in possible 28 GHz B-X-O experiments in B-X-O coupling measurements from spinning mirror EBE diagnostic (March 2006) will contribute to FY06 decision on technical feasibility of EBWCD on NSTX B-X-O coupling measurements from spinning mirror EBE diagnostic (March 2006) will contribute to FY06 decision on technical feasibility of EBWCD on NSTX

11EBW & HHFW Research - G. Taylor PAC-19 2/23/ GHz EBW Experiments on P EGASUS will Study Coupling, Propagation, Heating & CD Existing 2.45 GHz Existing 2.45 GHz PLT equipment to be PLT equipment to be used for 0.9 MW used for 0.9 MW EBW system EBW system Demonstrate EBW Demonstrate EBW coupling at significant coupling at significant power via O-X-B & X-B power via O-X-B & X-B Study nonlinear EBW Study nonlinear EBW coupling effects coupling effects Validate ray tracing, Validate ray tracing, demonstrate electron demonstrate electron heating heating Measure EBWCD Measure EBWCD

12EBW & HHFW Research - G. Taylor PAC-19 2/23/06 Frequency = 2.45 GHz, Poloidal Angle of Antenna Location = 15 deg. Initial modeling completed for P EGASUS 2.45 GHz EBW heating Initial modeling completed for P EGASUS 2.45 GHz EBW heating & CD system for I rod = kA & CD system for I rod = kA - CD efficiency ~ kA/MW ; - Fisch-Boozer CD dominates on axis for I rod = kA - I rod ~ kA may allow dominant off-axis Ohkawa CD kW EBW experiments could begin on P EGASUS in kW EBW experiments could begin on P EGASUS in Near Midplane EBW Launch Provides Significant EBW Electron Heating & Current Drive Near Axis

13EBW & HHFW Research - G. Taylor PAC-19 2/23/06 EBW Models & Emission Coupling Studies on NSTX Being Significantly Improved in FY06 NSTX & MAST EBW emission studies will contribute to FY06 technical NSTX & MAST EBW emission studies will contribute to FY06 technical assessment of EBWCD: assessment of EBWCD: - NSTX remotely steered antennas will study B-X-O coupling from L- & H-mode & effect of Li conditioning coupling from L- & H-mode & effect of Li conditioning - MAST spinning mirror diagnostic will study B-X-O coupling 2.45 GHz EBWCD modeling of P EGASUS plasmas at I rod > 200 kA to 2.45 GHz EBWCD modeling of P EGASUS plasmas at I rod > 200 kA to identify scenarios with dominant Ohkawa CD identify scenarios with dominant Ohkawa CD CQL3D/GENRAY modeling of EBE from non-thermal electron distributions CQL3D/GENRAY modeling of EBE from non-thermal electron distributions Include EBW mode conversion model in GENRAY Include EBW mode conversion model in GENRAY Benchmark CQL3D/GENRAY(with EBW MC) & BANDIT for NSTX cases Benchmark CQL3D/GENRAY(with EBW MC) & BANDIT for NSTX cases Improve user interface to "Prague" EBW emission modeling code Improve user interface to "Prague" EBW emission modeling code CQL3D/GENRAY EBWCD calculations assuming constant diffusion in CQL3D/GENRAY EBWCD calculations assuming constant diffusion in velocity space (consistent with TCV results) for a range NSTX conditions velocity space (consistent with TCV results) for a range NSTX conditions

14EBW & HHFW Research - G. Taylor PAC-19 2/23/06 EBW Priorities for FY Complete detailed study of EBE coupling with NSTX remotely steered antennasComplete detailed study of EBE coupling with NSTX remotely steered antennas Collaborate with 28 GHz startup/ramp-up experiments in Collaborate with 28 GHz startup/ramp-up experiments in & possibly 28 GHz B-X-O experiments in Upgrade radial transport calculation in the CQL3DUpgrade radial transport calculation in the CQL3D Examine possible role of self-consistent anti-pinch effect in Ohkawa EBWCD and its effect on bootstrap currentExamine possible role of self-consistent anti-pinch effect in Ohkawa EBWCD and its effect on bootstrap current Analytical study of EBWCD and scenarios for optimizing the current drive efficiencyAnalytical study of EBWCD and scenarios for optimizing the current drive efficiency Possible collaboration on NSTX ~1 MW, 28 GHz EBW experimentsPossible collaboration on NSTX ~1 MW, 28 GHz EBW experiments Implement ~ 400 kW, 2.45 GHz EBW heating & CD capability on P EGASUSImplement ~ 400 kW, 2.45 GHz EBW heating & CD capability on P EGASUS

15EBW & HHFW Research - G. Taylor PAC-19 2/23/06 NSTX HHFW Research HHFW research is directed towards… HHFW research is directed towards…

16EBW & HHFW Research - G. Taylor PAC-19 2/23/06 Field pitch angle may explain differences between co & counter k // (m -1 )% Power absorbed < 20 Parametric decay into surface waves may explain some of the power absorption dependence on k // HHFW Power Modulation Experiments Show Decreasing RF Absorption with Decreasing k //

17EBW & HHFW Research - G. Taylor PAC-19 2/23/06 Ion heating through parametric decay was observed in 2004 & confirmed with improved time resolution in 2005Ion heating through parametric decay was observed in 2004 & confirmed with improved time resolution in 2005 Waves in the surface can cause power deposition in sheaths and in the plasma periphery through collisionsWaves in the surface can cause power deposition in sheaths and in the plasma periphery through collisions For lower k //, fast waves propagate at lower density and reach much higher k  at a given density Wave fields in the edge are enhanced at lower k // and should contribute to greater power loss via parametric decay, sheath damping & collisions Wave fields in the edge are enhanced at lower k // and should contribute to greater power loss via parametric decay, sheath damping & collisions Possible HHFW Power Loss in Plasma Edge via Parametric Decay & Waves in the Periphery

18EBW & HHFW Research - G. Taylor PAC-19 2/23/06 Edge ion heating via parametric decay increases with wavelength: Edge ion heating via parametric decay increases with wavelength: k || = 14 m -1, k || = -7 m -1 Represents lower limit estimate of power loss by this process Represents lower limit estimate of power loss by this process Significant RF power required to sustain large temperature difference between edge ions and electrons Significant RF power required to sustain large temperature difference between edge ions and electrons Parametric Decay Accounts for Significant Fraction of RF Power Losses at k || = 14 m -1 & -7 m -1

19EBW & HHFW Research - G. Taylor PAC-19 2/23/06 k || = 3m -1 7m -1 14m -1 Density onset is function of ~ B*k || 2 Density onset is function of ~ B*k || 2 Propagation is very close Propagation is very close to wall at 7 m -1 and on the wall at 3 m -1 to wall at 7 m -1 and on the wall at 3 m -1 Losses in surface should be higher for lower k || Losses in surface should be higher for lower k || Strong motivation for antenna modification to Strong motivation for antenna modification to provide 14 m -1 directed launch provide 14 m -1 directed launch Fast Wave Propagation Begins at Lower Density at Lower k || B = 4.5kG kk 0 60 n e [x cm -3 ] 0 5

20EBW & HHFW Research - G. Taylor PAC-19 2/23/06 PDI instability observed at 4.5 kG, PDI should be weaker at higher magnetic fieldPDI instability observed at 4.5 kG, PDI should be weaker at higher magnetic field Onset density for propagation of HHFW is approximatelyOnset density for propagation of HHFW is approximately proportional to magnetic field at a given k || : proportional to magnetic field at a given k || : - waves propagate further from plasma edge at higher field Wave propagation into core faster with increased magnetic field:Wave propagation into core faster with increased magnetic field: - surface fields decrease with increasing magnetic field In 2006 will measure HHFW power loss properties as function of magnetic field at constant q:In 2006 will measure HHFW power loss properties as function of magnetic field at constant q: - New RF probes will detect waves propagating in periphery - Important for projections to the higher field regime of ITER & ST CTF; even 5-10% loss significant with ~ 20 MW ITER & ST CTF; even 5-10% loss significant with ~ 20 MW RF power RF power Higher Magnetic Field Should Yield Higher HHFW Heating Efficiency

21EBW & HHFW Research - G. Taylor PAC-19 2/23/06 New antenna voltage feedback algorithm to reduce RF power as voltage on antenna increases New antenna voltage feedback algorithm to reduce RF power as voltage on antenna increases This Year Early Solenoid-Free I P Ramp-up Experiments using HHFW Heating will Benefit from Antenna Voltage Feedback I P = 300 kA, k || = 7 m -1 co-CD I P = 250 kA, k || = 14 m -1 heating Time (s) Need to avoid RF tripping for 14 m -1 to get relaxation Need to avoid RF tripping for 14 m -1 to get relaxation 14 m -1 shows robust H-mode 14 m -1 shows robust H-mode 7 m -1 does not 7 m -1 does not

22EBW & HHFW Research - G. Taylor PAC-19 2/23/06 Magnetic field dependence of RF power loss, PDI ion heating and surface wave propagation & dampingMagnetic field dependence of RF power loss, PDI ion heating and surface wave propagation & damping If reversed B t & I p campaign, will conduct HHFW experiment to elucidate coupling physics using reversed fieldsIf reversed B t & I p campaign, will conduct HHFW experiment to elucidate coupling physics using reversed fields Effect of Li on HHFW antenna couplingEffect of Li on HHFW antenna coupling Non-inductive current ramp up using HHFWNon-inductive current ramp up using HHFW Heat CHI discharges with HHFWHeat CHI discharges with HHFW Early HHFW heating to reduce volt-seconds consumption in a long pulse dischargeEarly HHFW heating to reduce volt-seconds consumption in a long pulse discharge HHFW Experiments in 2006 Continue Exploring Coupling Physics & Application of HHFW to ST’s

23EBW & HHFW Research - G. Taylor PAC-19 2/23/06 HHFW Priorities for FY HHFW applications during current ramp-up phase of discharge:HHFW applications during current ramp-up phase of discharge: -Bridge gap between CHI initiation and NBI heating to high performance Improve understanding and performance of HHFW power coupling to the core plasma:Improve understanding and performance of HHFW power coupling to the core plasma: -Modify antenna to provide directed spectra at 14 m -1 for better CD efficiency (depending on results from FY2006 studies) and for higher power operation HHFW system will support driven CAE mode studiesHHFW system will support driven CAE mode studies Use improved coupler for HHFW-CD applicationsUse improved coupler for HHFW-CD applications