Transport Studies in Alcator C-Mod Catherine Fiore MIT Plasma Science and Fusion Center Transport Task Force March 26th Boulder, Co With Contributions from: I. Bespamyatnov, P. T. Bonoli*, D. Ernst*, M. J. Greenwald*, A. Ince-Cushman*, L. Lin*, E.S. Marmar*, R. McDermott*, M. Porkolab, M. Reinke*, J. E. Rice*, W. Rowan, S. Wukitch* *MIT-PSFC, FRC-UTA Supported by US DoE Alcator C-Mod
Introduction: s in Alcator C-Mod What s New: Gyrokinetic stability study at Improved ion temperature profile measurements Localization of fluctuation measurements Impurity transport at the foot Ohmic H-mode s LHCD effects
Features of C-Mod s Internal Transport Barriers (s) in C-Mod arise from steady H-mode plasmas lasting 2 or more energy confinement times when the central power input is low: they are seen in both Ohmic and Off-axis heated ICRF plasmas Alcator C-Mod 2 n e sqrt(z eff ) (1 /m 3 ) 12 1 8 6 4 t=1.3 s t=1.2 s Profiles peak t=1.1 s Transition t=1. s to t=.9 s t=.8 H-mode 2 ITb foot position.65.7.75.8.85.9 Major Radius (m) f ci min Moly tiles Divertor RF antenna
Features of C-Mod s C-Mod plasmas are a unique platform for study: No particle or momentum input Monotonic q profiles Collisionally coupled ions and electrons with T i T e Reduction in particle and thermal transport in the barrier region and core allows the Ware pinch to dominate the transport. This results in strongly peaked pressure and density profiles. Ion thermal transport is reduced to neoclassical Control of particle and impurity accumulation is achieved through application of central ICRF heating: TEM stability plays a role. Ernst, IAEA 24, 26
Increasing magnetic field moves ICRF resonance off-axis on low field side. s form when Bt > 6.2 T with ICRF at 8 Mhz R/L Te R/L Te 8 R=.78 m 4 (near barrier foot,inside) 5.4 5.6 5.8 6. 6.2 6.4 12 Toroidal Field (T) 8 R=.8 m 4 (barrier foot) 5.4 5.6 5.8 6. 6.2 6.4 Toroidal Field (T) R/L Te decreases in the region near the foot at the time of. Mrad/s.25.2.15.1.5. The region of stability to ITG modes widens with increasing magnetic field. ITG growth rate profiles.2.15.28.42.55 rho K. Zhurovich, et al, NP8.73 K. Zhurovich et al 27 Nucl. Fusion 47 122-1231.68.82 5.6 T 5.8 T 6. T 6.2 T 6.3 T
Upgraded high resolution x-ray spectrometer measurement of Ti profile shows barrier in temperature during Excellent agreement is found between T i from HIREX, central Ti from neutrons, and T e from Thomson Scattering. T i,t e (kev) 2. 1.5 1..5 Ti, HIREX Off-axis ICRF Ti, Neutron Inversion Te, Thomson Scattering..4.6.8 1. 1.2 1.4 1.6 t (s) Ti,Te (kev) 2. 1.5 1..5. t=.85, transition to t=.71 s, pre Ti, Neutron Inversion Te, Thomson Scattering Ti, HIREX Off-Axis ICRF t=1.9, foot region.7.75.8.85 Major radius (m) A temperature barrier was found previously with sawtooth heat pulse measurements and inferred from pressure profile increase Ion temperature profile from HIREX confirms T i transport barrier.
R/L Ti can be obtained from HIREX can be compared with that obtained from TRANSP R/L Ti from TRANSP increases from the center for all times; drop outside of core is not seen R/L Ti from TRANSP is comparable for core channels; shows discrepancy at larger radii. R/L Ti R/L Ti 15 1 5 15 1 5 1783128 foot..2.4.6.8 r/a.6.8 1. 1.2 time (s) Transp t=.7s H-mode t=.9s t=1.1s HIREX Data t=.71 s t=.85 s t=1.9 s Transp r/a=.2 core r/a=.4 near foot r/a=.6 Outer HIREX Data r/a=.16 r/a=.49 r/a=.58
Fluctuations arising during density peaking propagate in electron diamagnetic direction (Lin, P12) 1 8 6 4 Freq (khz)1 2 pci Fluctuations.9 1. 1.1 1.2 1.3 1.4 2.5 2. 1.5 density peaking 2..9 1. 1.1 1.2 1.3 1.4 time (s).4.3.2.1. Caveat: Neither non-linear GYRO nor linear GS2 has predicted TEM fluctuations in this plasma. Auto power [a.u.] B PCI TEM-like :TEM(GS2) :QC shot:1783121 t:[1.4,1.1]s f:[2,6]khz PCI was configured to preferentially view half of the plasma The fluctuations in 2-6 khz propagate in the electron diamagnetic direction e direction Wavenumber [cm-1]
Measurement of Boron impurity behavior in C-Mod s have been obtained with the CXRS diagnostic (Rowan, Bespamyatnov, Fiore: P18) Here is shown a simulation of B +5 density (solid line) and B +4 density (dash-dot line) compared with the measured B +5 density ( ). v/d (dashed line) required in the simulation is plotted on the right axis. It is found that light impurity confinement is improved in the. Light impurities accumulate in the region but transport does not reach neoclassical levels.
Density peaking in an begins shortly after an EDA H- mode develops for both Ohmic and Off-Axis ICRF s Density (1 2 /m 3 ) Density (1 2 /m 3 ) 4 3 2 1.65.7.75.8.85.9 Major Radius (m) 5 Ohmic H-Mode 4 3 2 1 Off-Axis ICRF Heated t=1.29 s t=.89 s H-mode t=.79 s L-mode t=.89 s H-mode t=1.14 s t=1.4 s } t=.79 s L-mode Density contour plot shows that the peaking time and intensity are similar Rmid, m Rmid, m.9.85.8.75.9.85.8.75 Electron Density, Shot 1783124 1 1 2 2 1 2 3 1 2.2.4.6.8 1. 1.2 1.4 t (s) Electron Density, Shot 1762819 2 1 2 4 1 2 6 1 2.65.7.75.8.85.9 Major Radius (m).7.8.9 1. 1.1 1.2 1.3 t (s)
Thermal transport characteristics are similar for off-axis RF induced and Ohmic H-mode s χ eff Ohmic 1763119 χ eff Off-axis ICRF 1783124 1.4 1.4 1.2 1.2 1. 1. 2 m /sec.8.6.4.2 r/a=.2 r/a=.4 r/a=.6 L-H transition Back transition 2 m /sec.8.6.4.2 r/a=.2 r/a=.4 r/a=.6 L-H transition.6.8 1. 1.2 1.4 t (s).6.8 1. 1.2 1.4 t (s)
R/L T appears to have different trend in Ohmic H- mode than in off-axis ICRF R/L Ti R/L Te 1 Ohmic H-mode, Transp 8 r/a 6 H-mode.6 4.4 2.2.6.8 1. 1.2 1.4 time (s) Ohmic H-mode, Transp 15 1 5.6 H-mode r/a.6.4.2.8 1. time (s) 1.2 1.4 R/L Ti R/L Te Off-axis ICRF, Transp 15 H-mode r/a 1.6 5.4.2.6.8 1. 1.2 1.4 time (s) Off-axis ICRF, Transp 15 r/a H-mode 1.6 5.4.2.6.8 1. 1.2 1.4 time (s)
R/L T in the barrier region dips as H-mode forms in Ohmic plasmas with s R/L Ti 1 8 6 4 No H-mode 1831321 1831332 2 r/a=.4 H-mode.6.8 1. 1.2 1.4 time(s) Ohmic H-mode plasmas from the same day are compared. One developed an and the other did not. R/LT i,e drops in the barrier region just after the H- mode forms in the case that generates an, and recovers to a higher level after the. The discharge that did not develop an had higher current, temperature, and toroidal field 2.5 2. 1.5 1. 2.8 2.4 2. Electron Temperature (kev) H-mode Density Peaking H-mode H-mode H-mode No 1831332 1831321 No 1831332 1831321 1.6.6.8 1. 1.2 1.4 time(s)
Ti (kev) Normalized temperature gradients obtained from high resolution x-ray measurement (R/L Ti ) show same trend with time in Ohmic H-mode s as that seen from TRANSP calculation 1.2 1..8.6.4.2. 1762814 Ohmic H-Mode 1.21s L-mode.71s 1.3s H-mode 1.5s 1.7s 1.9s Data was limited to r/a <.5, so Ti barrier is not resolved...2.4.6.8 rho.6.8 1. 1.2 1.4 1.6 time (s) Measured R/L Ti decreases with time after plasma enters H- mode. It rises with and as grows in the barrier region. R/L Ti 8 1762814 6 4 2 H-Mode Normalized Ti gradient from x-ray spectrometer r/a=.38 r/a=.29 r/a=.2 H-Mode and end
Is there an with LHCD? Counter toroidal rotation develops during LHCD, suggesting barrier formation. Central pressure increases. χ i decreases slightly in core with LHCD off-axis. More power is needed to obtain stronger effect for definitive analysis χ, Neoclassical χ i i Z-corrected Chang-Hinton χ, Neoclassical χ i i Z-corrected Chang-Hinton During LH Before LH injection χ i 1821514 +6 phasing χ neo 1821514 χ 182155 i -6 phasing χ 182155 neo
Conclusions Non-linear gyrokinetic modeling shows that broadening of the temperature profile with off-axis ICRF injection increases the size of the core-stable region of the plasma, reducing the outgoing particle flux, allowing Ware pinch to dominate particle transport. Improved temperature profiles establish that a thermal barrier exists at or near the particle barrier seen in the density profile. These profiles will improve data for stability analysis. Density fluctuations that arise and strengthen during growth propagate in the electron direction. However, gyrokinetic modeling does not find TEM in the weak case that shows this fluctuation. We will be working to reproduce strong and to obtain localized PCI measurements of this instability.
Conclusions (cont.) Light impurity confinement is improved in the. Light impurities accumulate in the region but transport does not reach neoclassical levels. s arising in Ohmic H-mode plasmas appear identical to those forming in off-axis ICRF heated cases. R/L T decreases following H- mode indicating that the trigger is likely the same mechanism as that determined for the off-axis ICRF s. Stability analysis is needed. LHCD plasmas show development of counter current rotation (See Ince-Cushman, et al. this morning) Slight temperature peaking suggests possibility of development.
PCI sees fluctuations arising during density peaking also in Ohmic H-mode s PCI spectrogram 1831321 PCI spectrogram 1831332 Frequency, Khz Frequency, Khz time (s) Hmode; starts 1.7 s, continues until end of shot time (s) H-Mode, No forms Chordal view, no masking