1. Modification of Edge Profiles, Edge Transport, and ELM Stability with Lithium in NSTX Rajesh Maingi, D. Boyle, J.M. Canik, J. Manickam, T.H. Osborne, P.B. Snyder, R.E. Bell, S.P. Gerhardt, R. Kaita, H.W. Kugel, B.P. LeBlanc, D.K. Mansfield, S.A. Sabbagh, and the NSTX Research Team 23 rd IAEA FEC Conference Daejeon, Rep. of Korea 11-16 Oct 2010 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 NSTX Supported by

2. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 2 Lithium wall coatings control recycling and edge density, and lead to ELM suppression End points of a well-controlled lithium coating sequence in which ELMs gradually disappear –Edge density, temperature, rotation and pressure profiles are modified with lithium –Edge cross-field D and  e drop from  N ~ 0.8-0.95 Edge peak pressure gradient moves farther from separatrix, and pedestal gets wider –Calculated bootstrap current changed, and edge stability improved

3. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 3 Edge localized modes (ELMs) observed in many non-lithium NSTX H-mode discharges D  [au] W MHD [kJ] Type I Type III Type V #108015 #112525 #111543 R. Maingi, JNM 2005  W/W MHD ~ 3-20%  W/W MHD ~ 1-5%  W/W MHD < 1%

4. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 4 Type I ELMs eliminated by lithium wall coatings Without-Li, With Li Lower NBI to avoid  limit Lower n e Similar stored energy H-factor 40% Higher P rad /P heat ELM-free, reduced divertor recycling H. Kugel PoP 2008 R. Maingi PRL 2009 M. Bell PPCF 2009 Predicted* by L. Zakharov in 2005 LITER Canisters ~ 700mg Li before 129038 * L. Zakharov, JNM 2007 I p [MA] P NBI [MW] n e [10 19 m -3 ] W MHD [kJ] H97L P rad [MW] D  [au]

5. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 5 T e, T i increased and edge n e decreased with lithium coatings No lithium With lithium separatrix R. Kaita IAEA 2008 D. Mansfield JNM 2009

6. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 6 Edge pressure profile modified with lithium No lithium With lithium (multiple time slices) separatrix

7. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 7 Profiles in discharges with lithium coatings only reproduced with recycling and transport reduction J. Canik, PSI 2010 R p =0.98, P loss =3.7 MW R p =0.92, P loss =1.9 MW (2D Plasma Neutral Transport Modeling With SOLPS) R P =divertor recycling coefficient Core SOL

8. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 8 Kinetic equilibrium fits using multiple data time slices integral to analysis procedure Profile fitting of multiple time slices (mapped to  N ) with standard procedures used as target for kinetic EFITs –Pre-lithium discharge profiles from last 20% of ELM cycle –Post-lithium discharge profiles used in 100-200 msec windows Free boundary kinetic EFITs run to match kinetic pressure profiles –Edge bootstrap current computed from Sauter neoclassical model No direct measurement biggest uncertainty –Stability evaluated with PEST code Fixed boundary kinetic EFITs run with variations of edge pressure gradient and edge current –Stability boundary evaluated with ELITE code

9. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 9 Peak edge pressure gradient and bootstrap current farther from separatrix with lithium coatings 129015, 0.4s 129015-20, 0.4 s 129038, 0.4 s 129038, 0.65s dP/d  N [kPa/  N ] NN J. B 0 0.3 0.6 NN NN Without-li With-li

10. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 10 Pre-lithium edge profiles close to kink/peeling instability threshold (ELITE) No lithium Low n=1-5 pre-cursor oscillations observed before ELM crash Mode growth rates low - why not diamagnetically stabilized? R. Maingi, PRL 2009 3 4 5 6 7 8 Normalized Pressure Gradient (  ) 1.2 0.8 0.4 0.1 0.05 Edge current [(jmax+jsep)/2 ]  lin /(  */2)

11. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 11 Edge profiles with lithium farther from instability threshold (ELITE) No lithium With lithium Relative roles of wider pedestals, current farther from separatrix, and magnetic shear being assessed R. Maingi, PRL 2009 3 4 5 6 7 8 Normalized Pressure Gradient (  ) 1.2 0.8 0.4 0.1 0.05 Edge current [(jmax+jsep)/2 ] 3 4 5 6 7 8 Normalized Pressure Gradient (  ) 1.2 0.8 0.4 0.1 0.05  lin /(  */2)

12. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 12 Density profile modification to lithium pumping the key in changing edge stability Lithium reduces recycling Core fueling reduced and edge barrier extended Density gradient reduced Peak edge P’ moves farther from separatrix, and pressure profile width broadens Edge bootstrap and parallel current modified Kink/peeling stability improved -> ELMs suppressed

13. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 13 ELMs disappeared gradually during experiment Reference (no lithium) With lithium Divertor D 

14. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 14 ELMy and ELM-free Discharges Ordered by Profile Broadening: Threshold for ELM stabilization No Lithium

15. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 15 Lithium wall coatings reduce edge transport and improve ELM stability in NSTX ELM-free phases increase gradually with lithium deposition, with discharges eventually becoming ELM-free –n e profile gradient reduced with increasing lithium –Edge T e, T i increase and profiles change substantially H-factor increased up to 50% for thickest lithium coatings –Region of low D,  eff extends inward from H-mode barrier –Global stability limits (  N ~ 5.5-6) encountered before edge (ELM) stability limits Peak pressure gradients shifted inward -> ELMs suppressed –Density profile modification crucial step toward ELM suppression Impurities accumulate and radiated power increases monotonically in the discharge  Present remedy: use 3d fields to trigger ELMs to purge impurities while looking to reduce impurity influx, e.g. via ‘snowflake’ divertor

16. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 16 EPH-mode phase observed for several  E, up to ~ 300 msec EPH P NBI /10 [MW] EP H-mode H-mode separatrix t 1 t 2

17. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 17 THANK YOU FOR YOUR ATTENTION

18. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 18 BACKUP

19. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 19 3D external fields used to trigger ELMs, prevent radiation buildup while keeping high energy confinement from lithium Type I ELMs triggered for impurity control “Snowflake” divertor reduced (post-lithium, n=3)impurities J. Canik, PRL 2010, also this conf. V. Soukhanovskii, this conf.

20. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 Page 20 Comparison of Standard and EP H-mode profiles

21. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 21 Details of ELM suppression and methods to reduce impurity buildup under active research Several outstanding questions: –Why are the ELMs not stabilized by diamagnetic drift, as in higher aspect ratio tokamaks? Low growth rates ~  lin /  A > 1% unstable experimentally Should be stabilized by diamagnetic drift:  lin /(  */2) < 5-10% –Complete evolution: why do ELMs go away the way they do i.e. with increasing periods of quiescence? Details of density/pressure profile modification may be beyond present ability to measure experimentally –Additional Thomson channels being installed for 2011 –Does modification to v  profile play a role?

22. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 22 3D external fields used to trigger ELMs, prevent radiation buildup while keeping high energy confinement from lithium Type I ELMs triggered for impurity control Edge T e and dT e /dr increased (post-lithium, n=3)--> n=3 more unstable (PEST) Reference n=3 applied J. Canik, PRL 2010

23. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 23 Quiescent phases ( * ) increase with increasing lithium coating (P NBI = 4 MW) * 129019 129021 129023 129027 129030 129038 (P NBI = 2 MW) No lithium Increasing lithium coating D. Mansfield, JNM 2009 * * * * * * * * Div. D 

24. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 24 NSTX Facility Capabilities R, a max 0.85, 0.67 m Aspect ratio A1.27 – 1.6 Elongation  1.6 – 3.0 Triangularity  0.3 – 0.8 Toroidal Field B T0 0.3 – 0.55 T Plasma Current I p ≤ 1.5 MA Auxiliary heating: NBI (100kV)≤ 7.4 MW RF (30MHz)≤ 6 MW Central temperature1 – 6 keV Central density≤1.2  10 20 m -3 Graphite/CFC PFCs + Lithium coating Slim center column with TF, OH coils Excellent diagnostic access

25. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 25 Confinement improves with lithium coatings, due to broadening of the temperature profiles TRANSP analysis confirms electron thermal transport in outer region progressively reduced by lithium M. Bell PPCF 2009 S. Ding PPCF 2010

26. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 NSTX Developing Lithium-Coated Plasma Facing Components (PFCs) 26 2005:Lithium pellet injection for wall coatings 2006:LIThium EvaporatoR (LITER) deposited lithium on center column and lower divertor 2007:Larger evaporator re-aimed to increase deposition rate on lower divertor 2008:Dual LITERs to eliminate shadowed regions –Also used “lithium powder dropper” 2009:Routine use of dual LITERs 80% of discharges now have lithium applied beforehand Complements and builds on experience with lithium coating of limiters in tokamaks TFTR, CDX-U (liquid), T-11, FTU, HT-7, TJ-II LITER Canisters

27. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 27 Global  N limit encountered before edge stability limit with lithium coatings Pre-Li, Post-li, Post-li at  limit Intermediate NBI to probe  limit  N limit ~ 5.5 with P NBI =3 MW R. Maingi, PRL 2009 I p [MA] P NBI [MW] n e [10 19 m -3 ] W MHD [kJ] H97L P rad [MW] D  [au]

28. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 28 ELM evolution with shot number Reference (no lithium) With lithium

29. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 29 ELM evolution with shot number With lithium Ohmic Locked Mode No PFs, TF Locked Mode Higher fueling, lower NBI

30. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 30 SOLPS modeling used to model power and particle balance of baseline ELMy discharge J. Canik, PSI2010 Pre-li Data Model R p =0.98, P=3.7 MW Core SOL

31. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 31 Post-lithium discharge profiles not reproduced with simple recycling coefficient change Post-li data R p =0.98, P=3.7 MW R p =0.90, P=1.9 MW J. Canik Core SOL

32. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 32 Post-lithium discharge profiles better matched with transport and recycling coefficient change J. Canik Pre-li Data Model R p =0.92, P=1.9 MW Core SOL

33. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 33 Electron pressure gradient dominates total pressure gradient nDnD PiPi P i+e No lithium 129015-21, 0.4 s NN NN

34. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 34 Electron pressure gradient dominates total pressure gradient nDnD PiPi P i+e With lithium 129038, 0.4 s NN NN

35. NSTX IAEA FEC 2010 meeting: ELM suppression with lithium R. Maingi EXC/2-211-16 Oct. 2010 35 Measured edge bootstrap current in reasonable agreement with neoclassical calculation in DIII-D D. Thomas, PRL 2004