Bulletin of the American Physical Society
50th Annual Meeting of the Division of Plasma Physics
Volume 53, Number 14
Monday–Friday, November 17–21, 2008; Dallas, Texas
Session NI1: Edge and Pedestal Physics in Tokamaks |
Hide Abstracts |
Chair: Earl Marmar, MIT Room: Landmark A |
Wednesday, November 19, 2008 9:45AM - 10:15AM |
NI1.00001: Development and Validation of a Predictive Model for the Pedestal Height Invited Speaker: P.B. Snyder The pressure at the top of the edge transport barrier (or ``pedestal height") strongly impacts fusion performance. Predicting the pedestal height in future devices such as ITER remains an important challenge for plasma theory. While uncertainties remain, MHD stability calculations, accounting for diamagnetic stabilization, have been largely successful in predicting the observed maximum pedestal height, when the barrier width is taken as an input. However, the strong correlation between the stability-constrained height and the width, along with measurement uncertainty in the width, have complicated prior efforts to discern the dependencies of the width. Here, we employ the peeling-ballooning stability calculations as a constraint, accounting for the strong correlation of the width and height, and allowing study of the dependencies of the width itself. A simple equation for the pedestal width is derived [$\sim (\beta_{pol\, ped})^{1/2}$] and successfully compared to a large set of DIII-D data. Combining this simple model of the pedestal width with direct stability calculations using the ELITE code, we develop a new predictive pedestal model, EPED1, which allows quantitative prediction of the pedestal height and width in both past and future experiments. Pedestal height predictions with EPED1 were made before a dedicated set of pedestal height variation experiments on DIII-D. The predictions were found to be in very good agreement with the DIII-D observations, in which pedestal height varied more than an order of magnitude. EPED1 is found to quantitatively capture the observed complex dependencies of pedestal height on shape, $q$, collisionality and global beta in large statistical studies. We present comparisons to observations on multiple tokamaks, as well as predictions for ITER and other next step devices. [Preview Abstract] |
Wednesday, November 19, 2008 10:15AM - 10:45AM |
NI1.00002: Edge Radial Electric Field Structure on Alcator C-Mod and its Connection to H-Mode Confinement Invited Speaker: R.M. McDermott High-resolution charge-exchange recombination spectroscopy measurements using B$^{+5}$ ions have enabled the first calculations of the radial electric field in the C-Mod edge. These observations, made in discharges with no externally applied torque, provide for important comparisons with other devices and offer new challenges for theory and simulation. Qualitatively, the field structure on C-Mod, computed from the B$^{+5}$ radial force balance equation, is similar to that observed on other tokamaks. However, the depths of the C-Mod E$_{r}$ wells (up to 300kV/m) are unprecedented -- over twice as deep as on other devices and the narrow well widths ($\sim $5mm) suggest a scaling with machine size. The poloidal velocity is found to be the dominant term in the force balance equation contributing up to 200kV/m transiently after L-H transitions and up to 80kV/m during steady H-modes. This differs from measurements on other tokamaks, in which the diamagnetic term is dominant. The radial electric field during EDA H-modes is steady in time, while a clear evolution of the E$_{r}$ well depth is observed in ELM-free H-modes. The depth of the well decays as the electron temperature pedestal height (T$_{ped})$ decreases due to increases in P$_{rad}$. Interestingly, it is the poloidal VxB term, not the diamagnetic term, which decays. In fact, the diamagnetic term is observed to make a fixed contribution to E$_{r}$ independent of confinement and pedestal heights, while the poloidal velocity contribution scales with T$_{ped}$, qualitatively consistent with the neo-classical formulation that poloidal impurity flows are driven in part by ion temperature gradients. As a further important observation, we have compared E$_{r}$ with an estimate of the main ion diamagnetic term and found that they are roughly equal, indicating that the VxB term for the main ions is small. [Preview Abstract] |
Wednesday, November 19, 2008 10:45AM - 11:15AM |
NI1.00003: Edge Pedestal Control in Quiescent H-Mode Discharges in DIII-D Using Co Plus Counter Neutral Beam Injection Invited Speaker: K.H. Burrell Active control of the edge pedestal transport has been achieved in quiescent H-mode plasmas by using co plus counter neutral beam injection (NBI) to modify the edge toroidal rotation speed. This allows continuous adjustment of the pedestal density and pressure by a factor of $\sim$2 while avoiding edge localized modes (ELMs), thus permitting operation near but below the ELM stability boundary. These plasmas exhibit edge particle transport more rapid than that produced by ELMs while operating at reactor relevant pedestal beta $\sim$1\% and collisionality $\sim$0.1; pedestal densities up to 1/2 the Greenwald density have been achieved. The essential feature distinguishing QH-mode from standard ELMing H-mode is the presence of an edge-localized electromagnetic mode, the edge harmonic oscillation (EHO). The EHO provides sufficient particle transport to maintain the edge pressure below the ELM stability boundary. The EHO is spontaneously generated by the plasma itself and requires no external coils to generate a perturbed magnetic field. Experimental observations suggest rotation controls the edge pedestal by altering the EHO-induced particle transport. Calculations using the ELITE code show that the QH-mode operating point is near the peeling stability boundary. Much of the physics of the EHO is consistent with a model in which the EHO is an edge kink-peeling mode destabilized by shear in the edge toroidal rotation at an edge current density slightly below that on the standard ELM boundary. This theory predicts that, given sufficient rotational shear, QH-mode should be achievable with either counter or co rotation. QH-mode was first discovered in counter rotating plasmas; we now have seen QH-mode periods up to 1 second long in co-rotating plasmas. [Preview Abstract] |
Wednesday, November 19, 2008 11:15AM - 11:45AM |
NI1.00004: Finite drift orbit effects in a tokamak pedestal Invited Speaker: Grigory Kagan Experiments show that the pedestal width is typically of the order of the poloidal gyroradius. Therefore, an analytic pedestal model should treat physics on this scale. It is also desirable to include finite Larmor radius (FLR) effects. To keep these features we develop a special version of gyrokinetics that employs canonical angular momentum as one of the variables. This technique encompasses both of the preceding scales in an uncoupled manner. In particular, it allows considering the limit in which FLR effects are neglected, while neoclassical and zonal flow phenomena are retained. Using our gyrokinetic equation in this limit we find that the background ion temperature profile cannot vary on the scale of the poloidal gyroradius. Pressure balance analysis in the pedestal for subsonic ion flow then yields that the ions in the pedestal are predominately electrostatically confined, giving a strong pedestal radial electric field consistent with recent C-Mod measurements. As a result, particle trajectories in the pedestal are very different from those in the core and zonal flow and neoclassical phenomena are altered. Therefore, we are led to consider single particle motion in a tokamak in the presence of a strong radial electric field and its impact on collisionless zonal flow in pedestal. For instance, we find that due to large ExB, the neoclassical polarization response of trapped particles is different from that in the Rosenbluth-Hinton (RH) case. To treat this problem kinetically our version of gyrokinetics turns out to be convenient. Using it we obtain an analytical expression for the zonal flow response. Interestingly, our analysis predicts a spatial phase shift in the residual relating the final and initial electrostatic potential level of the zonal flow - an effect absent in the RH case. [Preview Abstract] |
Wednesday, November 19, 2008 11:45AM - 12:15PM |
NI1.00005: Radially compressed full-f ITG turbulence dynamics across the pre-transition L-mode edge pedestal in magnetic separatrix geometry Invited Speaker: Choong-Seock Chang We find from the full-f XGC1 gyrokinetic edge code that there is ITG turbulence excited in a real-geometry, pre-transition L-mode edge pedestal that includes a separatrix and X-point. The neoclassical and turbulence dynamics are simulated together in numerical g-eqdsk magnetic equilibrium with material wall. It has been a conventional assumption in the L-H transition scenarios that the ITG mode itself is stable in the L-mode edge since $\eta _{i}$ in the pedestal stays below the conventional critical $\eta _{ic}$ for the linear ITG instability onset due to broader T$_{i}$ profile than the density profile. Since the broader T$_{i}$ profile yields high $\eta _{i} \quad > \quad \eta _{ic}$ at the density pedestal top (and radially inward), the ITG turbulence activity has been assumed to be limited to the core side in from the pedestal. The strong ExB shearing in the pre-transition pedestal was supposed to prevent the turbulence spreading. Surprisingly, we find that the turbulence grows simultaneously from the pedestal top to the bottom. Unlike in a core plasma, the growth of the radial streamers is accompanied with zonal flow growth from the beginning, indicating that the ITG mode growth is quasilinear in the edge. This observation suggests a new possibility for the understanding of L-H transition mechanism. The ion thermal conductivity $\chi _{i}$ is found to be in the range of experimental observations. Turbulence is severely compressed in the pedestal toward the separatrix, implying a global interaction of the turbulence with the many spatially coupled nonlocal edge-specific kinetic mechanisms, including the strong neoclassical ExB and the Reynolds stress mean field in the high-q separatrix region (with X-point), strong plasma gradient, finite ion orbit dynamics with transitional collisionality, and the nonlocal GAM energy exchange with turbulence. [Preview Abstract] |
Wednesday, November 19, 2008 12:15PM - 12:45PM |
NI1.00006: Gyrokinetic turbulence under near-separatrix or non-axisymmetric conditions Invited Speaker: Frank Jenko The comprehensive gyrokinetic turbulence code GENE can use geometric information extracted from MHD equilibria, including the near-separatrix regions of tokamaks and non-axisymmetric equilibria of stellarators. This allows comparisons between simulations of microturbulence in the core versus the edge and in tokamaks versus stellarators. The GENE simulations can include the interactions among ion temperature gradient (ITG), trapped electron mode (TEM), and electron temperature gradient (ETG) turbulence. Simulations of the tokamak core show a strong nonlinear interaction among these modes, which modifies the transport. Moreover, multiscale simulations show that for realistic ion heat (and particle) flux levels, sub-ion scales generally have to be included in comprehensive transport models. In the very edge region of a tokamak, the turbulence also tends to acquire a multi-scale nature, being driven, e.g., by ETG modes peaking near the X-point and by microtearing modes. Nonlinear simulations will be presented which help characterize the residual anomalous transport in the H-mode edge. Large differences are seen in the criteria for marginal stability, the rapidity of the increase in the level of transport above marginality, and the importance of zonal flows when GENE is used to study ITG microturbulence in several tokamaks and in various optimized stellarators. This allows an assessment of the potential for modifying and optimizing mictroturbulence by non-axisymmetric shaping. [Preview Abstract] |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2020 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
1 Research Road, Ridge, NY 11961-2701
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700