Impurity confinement and transport in high confinement regimes without edge localized modes on DIII-Da)

Abstract

Impurity transport in the DIII-D tokamak [J. L. Luxon, Nucl. Fusion 42, 614 (2002)] is investigated in stationary high confinement (H-mode) regimes without edge localized modes (ELMs). In plasmas maintained by resonant magnetic perturbation (RMP), ELM-suppression, and QH-mode, the confinement time of fluorine (Z ¼ 9) is equivalent to that in ELMing discharges with 40 Hz ELMs. For selected discharges with impurity injection, the impurity particle confinement time compared to the energy confinement time is in the range of s p =s e % 2 À 3. In QH-mode operation, the impurity confinement time is shown to be smaller for intense, coherent magnetic, and density fluctuations of the edge harmonic oscillation than weaker fluctuations. Transport coefficients are derived from the time evolution of the impurity density profile and compared to neoclassical and turbulent transport models NEO and TGLF. Neoclassical transport of fluorine is found to be small compared to the experimental values. In the ELMing and RMP ELM-suppressed plasma, the impurity transport is affected by the presence of tearing modes. For radii larger than the mode radius, the TGLF diffusion coefficient is smaller than the experimental value by a factor of 2-3, while the convective velocity is within error estimates. Low levels of diffusion are observed for radii smaller than the tearing mode radius. In the QH-mode plasma investigated, the TGLF diffusion coefficient is higher inside of q ¼ 0:4 and lower outside of 0.4 than the experiment, and the TGLF convective velocity is more negative by a factor of approximately 1.7.

MHD modeling of a DIII-D low-torque QH-mode discharge and comparison to observations

“…8 of Ref. [16] of 2500 m as τ cond, L 2 < n 0 > χ n 0 (ψ 0 ) 1.5 × 10 −2 s, (6) which is longer than the convective confinement time. The factors of density are added to produce an evaluation of the parallel thermal conduction in the pedestal that is consistent with our temperatureevolution equation (Eqn.…”

The quiescent H-mode regime for high performance edge localized mode-stable operation in future burning plasmasa)

“…In these experiments, a fluorine impurity was injected (as carbon tetrafluoride, CF 4 ) in plasmas of similar density, but different types of H-mode edge: ELMing, RMP ELM-suppressed, and QH-mode. 32 Fluorine was chosen because it is not normally present in the tokamak environment (non-intrinsic) and it is highly reactive, so that once it exits the plasma core it does not re-enter (non 19. Poincar e section of magnetic field lines near the divertor in the JOREK simulation with saturated kink-peeling modes (n ¼ 1 À 5).…”

“…It does, however, depend on the equilibrium density, and the magnitude of the EHO amplitude. 32 Results of the dependence on the atomic number of the impurity ion are being analyzed.…”

Light impurity transport in JET ILW L-mode plasmas

“…Despite all the efforts, some differences are still present between the predicted peaking of light impurities and the experimental values. In JET [3,5,8] and ASDEX Upgrade [3] H-mode plasmas, differences between the experimental density peaking of 3 He, C, B and N i and the predicted peaking by gyro-kinetic and neoclassical theory have been found, while in DIII-D [17] H-mode plasmas, differences between density peaking of F and quasi-linear and neoclassical predictions have been observed, indicating that something may be missing for a complete understanding of the particle transport of light impurities. With respect to previous JET studies made in C-wall H-modes, this study was performed in JET ILW L-modes, increasing the number of light impurities studied, by measuring in the same discharges the profiles of 3 He, Be, C, N and N e with different heating schemes.…”