The current status of the Lithium beam diagnostic at ASDEX Upgrade Elisabeth Wolfrum Josef Schweinzer Matthias Reich Rainer Fischer Max Planck Institut für Plasmaphysik Garching, Germany
Outline Experimental setup Edge ion temperatures CX with He 2+ CX with D + Edge ion densities Edge electron densities ELM resolved profiles LID evaluation with Bayes integrated concept Wide Filters: high temporal resolution
Experimental setup Beam emission spectroscopy Electron density measurement Li 0 (2p-2s) @ 670.8 nm LID Li-beam: 30 80 kev, 2-4 ma 12 mm Charge exchange spectroscopy Ion temperature measurement M.Reich - Thesis Ion density measurement LIT LIS
EM CCD improves availability of T i measurements. counts counts 15 10 5 0-5 150 100 50 0-50 PI Micromax CCD camera He II (4-3) 468.0 468.6 469.2 λ(nm) PhotonMax EM CCD camera He II (4-3) 468.0 469.0 470.0 λ(nm) Signal/noise ratio much improved due to EM technology Same radial position Measured temperatures agree, Ti = 370 ev Temporal resolution: 10 channels (LOS) on one CCD with 4 ms continuously active CCD area frame transfer traditional amplifier 1 MHz, 5 MHz EM amplifier Extended serial register: 5 MHz, 10 MHz Electron multiplication via impact ionisation
Edge ion temperature profiles T i (kev) T i (kev) Ti (kev) Ti (kev) 4 3 2 1 2.5 2.0 1.5 1.0 CEZ core T i @ 5 s # 20214 Li-beam T i @ 4.5-5.7 s 0.0 0.2 0.4 0.6 0.8 1.0 1.2 rho poloidal rho poloidal # 20215 Li-beam T i @ 4.5-5.7 s AA:2 AB:2 AA AC Spatial resolution ~ 5 mm Temporal resolution not available: Signal must be integrated over 1-2 s. He concentration > 10%. L-mode o.k. H-mode: only for f ELM < 100 Hz. 0.5 CHZ core T i @ 5 s 0.0 0.2 0.4 rho 0.6 poloidal 0.8 1.0 1.2 rho poloidal
New: CX measurements also possible with D + ions. Raw data: spectrum temporal evolution of integrated spectrum CCD counts wavelength (nm) CCD counts counts (a.u.) time (s) net spectrum wavelength (nm) No ELMs or regular ELMs ELMs have to be cut out t > 500 ms integration time Fit difficult because centre is always dominated by photon statistics of passive line emission Inclusion of CXS_fit in progress
Edge ion densities, examples C 6+ density (m -3 ) 1 10 18 5 10 17 #18055, t = 3-5 s Core charge exchange Li-beam, ADAS Li-beam, no collisional mixing Li-beam, fully mixed 0 0.6 0.7 0.8 0.9 1.0 1.1 normalised poloidal radius C 6+ He 2+ temperature (ev) density (m -3 ) 600 500 400 300 200 100 0 1 10 20 1 10 19 #20463, t = 3.9-5 s electrons ions electrons He 2+ 1 10 18 0.85 0.9 0.95 1.0 1.05 1.1 normalised poloidal radius
Beam emission spectroscopy BES 0 Li + plasma Li(2p 2s)@ 670. 8nm Lithium beam attenuation code: dni ( z) = dz [ n ( z) a ( T( z)) + b ] N ( z) e ij N i relative occupation of state i (i=2s, 2p, 4f, Li + ) a ij rate coefficients b ij Einstein coefficients ij j 4s 497.2 nm 4d 3d 460.3 nm a ij : Inelastic collisions with protons, electrons and impurities b ij : radiative transitions References: Schweinzer et al: At.Data Nucl Data Tables 72 (1999) 239-273 Brandenburg et al: PPCF 31 (1999) 471-484 610.4 nm 2p 670.8 nm 2s Li I Grotrian diagram
Electron density measurements Measured profile + errors Produce fit to data Li I This relative profile Li 2p (z) is directly related to occupation number of Li(2p). αli 2p (z) = N 2p (z), α = const. z (cm) α is determined via 2 boundary conditions: N i (z=0) = δ 1i N 1 (z end ) = 0 dn ( z) dz [ ] i Use second equation of = n ( z) a ( T( z)) + b N ( z) e ij Lithium beam attenuation code ij j to get n e. Singularity near maximum of Li(2p) profile.
ELM resolved with int times ~ 1ms 10 8 Time interval extended to include Raus scan: Shot: 12200 t1= 3.400 t2= 5.200 all available data 10 8 LID with 1 ms temp. resolution Shot: 12200 t1= 3.400 t2= 5.200 only data in window rel. to ELM: -3.5 ms to -1.5 ms Electron density [10**19 m**-3] 6 4 Electron density [10**19 m**-3] 6 4 2 2 Edge Thomson scattering Lithium beam Edge Thomson scattering Improved Lithium beam 0 0.90 0.95 1.00 1.05 1.10 1.15 rho-pol Printed by: epw Tue Jan 16 15:54:48 2007 0 0.90 0.95 1.00 1.05 1.10 1.15 rho-pol Printed by: epw Tue Jan 16 16:25:06 2007
Binning of raw signal relative to ELM yields density profiles across ELM. 0.4 0.3 0.2 0.1 0.0 1.0 0.8 0.6 0.4 0.2 0.0-0.2 Dalpha:21160/AUGD/POT(0)/ELMa-Han LIB13:21160/AUGD/LIB(0)/SIG13 #21160 Hα LIB:SIG13 4.9 5.0 5.1 5.2 Time (s) Choose time interval with regular ELMs Determine for every t in LIB signal: t to previous ELM t to next ELM Add all signals with same temporal distance to ELM Calculate density profile Attention: ELMs should have about same size. ELM shotfile must be checked carefully: no missed or additional ELMs. Lithium beam must be very stable: no sparks.
n e gradient in ETB region is recovered after 3 ms. 10 20 #21160 n e (m-3) 10 19 10 18 n e (10 19 m -3 ) ne (e19) 6 5 4 3 2 1 t_elm (ms) -3-2 -1 0 +1 +2 +3 +4 10 17 5.000 4.995 4.990 4.9850 5 10 15 time (s) x (cm) 2.10 2.12 2.14 2.16 2.18 2.20 2.22 R(m) R(m) @ z=ma
Determine electron density profile given Measured data of Li I (2p- 2s) profile and their likelihoods! accurate error determination! Measurements of other diagnostics. Rainer Fischer: integrated concept Description of profile: 13 knots using Additional information: Hermite polynomials n e profile monotonic Lithium beam attenuation code Li I (2p-2s) profile Chi 2 fit to the data * factor α to determine 13 knots and α.
Electron density profile evaluation now beyond point of singularity. Electron density ne (e19) (1e19 m -3 ) 12 10 8 6 4 2 Li beam, new Li beam Thomson #20160 t = 3.8 s 0.80 0.85 0.90 0.95 1.00 1.05 1.10 1.15 rho poloidal AA AA AF Medium core density: Pedestal well determined. Temporal resolution: 1 ms So far LID density evaluation stops just before turn. New: high certainty of profile up to ρ pol = 0.93 cview-prof (gdc) v2.49 - User: epw - Fri Jan 19 11:21:35 2007 /afs/ipp/home/e/epw/cview/stdset/comprrf.cvp : 20160 Electron density ne (e19) (1e19 m -3 ) 12 10 8 6 4 2 Li beam, new Li beam Thomson #20160 t = 6.6 s 0.80 0.85 0.90 0.95 1.00 1.05 1.10 1.15 rho poloidal AA AA AF High core density: So far LID density evaluation stops in SOL. New: reliable profile up to ρ pol = 0.96
Latest improvement: broader filters #22287, 3-4s, t_exp = 4ms, LIA channel 4 He I Li I black=beam off red = beam on C II C II, C III New filters: 2 nm FWHM, before 0.5 nm Easier to change to different beam velocity (no tilt adjustment necessary) Higher transmission (85%, before 50%) Signal ~ factor 10 larger Now 1.5 4 sec with 20 khz (before 5 khz) Li I, unshifted Li I, blue shifted (upper optic) Allows faster data acquisition: 1.5 4 sec with 20 khz (before 5 khz)
Summary Ion temperature profiles No temporal resolution ( t > 500 ms) Good spatial resolution (5 mm) C: concentration now too low (< 0.4 %) He: good data if concentration > 10% D: good data if plasma quiet, e.g. ohmic or L-mode Ion density profiles Collisional mixing is important at the edge. Electron density profiles (main business!) Excellent temporal resolution (~ 1 ms). Li-beam electron densities can resolve ELM, if several light profiles are binned relative to ELM. Integrated concept: edge pedestal densities can be determined up to n e PED ~ 7 10 19 m -3. New filters give more photons, more flexibility and allow faster data acquisition (20 khz).