APS-DPP October Denver, Colorado CDX-U Final results from the CDX-U lithium program R. Majeski, R. Kaita, H. Kugel, T. Gray, D. Mansfield, J. Spaleta, J. Timberlake, L. Zakharov, Princeton Plasma Physics Laboratory, Princeton, NJ, USA V. Soukhanovskii, Lawrence Livermore National Laboratory, Livermore, CA, USA R. Maingi, Oak Ridge National Laboratory, Oak Ridge, TN, USA  High power density liquid lithium heat flux testing with an electron beam system  Recycling reduction with lithium coatings and a liquid lithium limiter

APS-DPP October Denver, Colorado CDX-U Three lithium, two gas fueling systems available R 0 =34 cm a = 22 cm  1.6 B T (0)2.2 kG I P  80 kA  disch <25 msec T e (0)~100 eV n e (0)<6x10 19 m -3 CDX-U:  Lithium tray limiter –300 g of lithium in a toroidal tray  Electron beam high heat flux, lithium coating system –Used lithium tray inventory as source  Resistively heated lithium evaporator –NSTX prototype  Gas injection systems –Wall mounted piezo valve –Supersonic gas injector R 0 = 34 cm Width = 10 cm 6 mm deep  Up to 1000Å of lithium coatings between discharges

APS-DPP October Denver, Colorado CDX-U High power density electron gun intended to “spot heat” lithium Charging of probe tip insulator disturbs beam Gaussian beam profile, width 3 mm  Converted commercial gun  4 kV, mA typ.  sec. run typical  Uncooled (Tantalum, Macor, SS)  Total power modest: <1.6 kW  Power density high: < 60 MW/m 2  Objective: 1000Å lithium wall coatings –TF + VF used to guide beam »Can be pulsed to 600G; typ. 200 G –Lithium tray fill (~3 mm deep)used as evaporation target. »Lithium area ~600 cm 2 >> beam spot  Spot heating proved impossible

APS-DPP October Denver, Colorado CDX-U Tray temperatures Lithium Beam cycles Temperature °C Time (h:mm) (1 hr)  Small beam spot heated entire tray of lithium solid lithium Melting point Melt wave reaches far thermocouple Beam spot

APS-DPP October Denver, Colorado CDX-U  Electron beam heating induces flow  Flow very effectively inhibits localized heating  IR camera movie of 25 sec. of a 300 sec. beam run  Yellow denotes +55°C, red denotes +110°C  Field ramps from 200 G to 400G 10 sec into clip  If only conduction were active, area under beam would heat to 1400ºC in 0.1 sec. Beam spot Entire tray heated, produced lithium wall coatings Framing pauses, white flag at field ramp  Localized heat deposition (and/or beam current) induces lithium flows –Marangoni effect; temperature- dependent surface tension IR image Visible image Centerstack Lithium in tray

APS-DPP October Denver, Colorado CDX-U Full wall coatings + liquid lithium produced very high particle pumping rates  Particle pumping rate in CDX-U is  part/sec.  Sufficient to pump a TFTR supershot –But the active wall area in CDX-U is only 0.4 m 2 –~Two orders of magnitude less than the active wall area in TFTR during lithium wall conditioning.  Liquid lithium also eliminated all traces of water –Oxygen vastly reduced  Carbon, other impurities also reduced  The effective particle confinement time in the presence of recycling,  p *   p /(1-R), is reduced dramatically with liquid lithium limiters and wall coatings –  p * too long to measure in the complete absence of lithium wall coatings p*p*

APS-DPP October Denver, Colorado CDX-U Ohmic confinement is increased  Magnetics have been extensively revamped  Equilibrium and Stability Code (ESC) rewritten to include effects of wall eddy currents (Zakharov)  Diamagnetic loop, reconstructions used for  E –Nonstationary equilibrium –Poynting flux calculation  Observed confinement times significantly exceed ITER98P scaling –ITER98P provides best fit to START data –START was similar in size to CDX-U

APS-DPP October Denver, Colorado CDX-U Increased confinement time is correlated with reduced recycling coefficient (R)  No direct measure of  p available  Here we assume that  p ~  E –If  p >  E, then we are overestimating the recycling coefficient –  p is unlikely to be much less than  E  Lowest global recycling coefficients ever recorded for a magnetically confined plasma –First plasma not dominated by wall fueling source  Range in line averaged density:  m -3 (prefill only: 6  m -3 )  Range in I p : kA (prefill only: 62 kA) Prefill only

APS-DPP October Denver, Colorado CDX-U Summary  Simple, 3 mm deep liquid lithium pool was very effective at redistributing extremely high power density heat loads (~50 MW/m 2, 300 s.) –Fields up to 600G ineffective at suppressing convective effects –Heat transport mechanism modeled as Marangoni effect (Zakharov) –Bodes well for liquid lithium PFCs  Simultaneous deployment of liquid lithium limiter Å between-shots lithium wall coatings  Particle removal rates produced in CDX-U sufficient to pump a TFTR supershot  Recycling coefficients of < 50% are the lowest ever achieved in a magnetically confined plasma  2-3  enhancement in low recycling discharge confinement times over pre- lithium case –Largest increase in ohmic tokamak confinement ever observed  CDX-U now being disassembled, converted to LTX –Full liquid lithium wall See posters RP (Kaita), RP (Gray), RP (Spaleta) Thursday afternoon