Session PO3: C-Mod Tokamak

Overview of Alcator C-Mod Research

C-Mod research has emphasized lower hybrid current profile control, realtime ICRF matching, pedestal studies including large ELMs, SOL turbulence and transport including flow studies, hydrogenic retention in metals, wall conditioning, disruption prediction and mitigation, error fields and locked modes, exploration of lower density and collisionality regimes, and Alfven eigenmodes. Particular attention is given to high priority R\&D for ITER and joint experiments coordinated through the ITPA. LHCD experiments have been extended to H-Mode regimes, and in combination with ICRF, with good coupling across the edge plasmas. An important new tool for C-Mod is active density control with a divertor crypump. Improved measurements across all plasma regimes are enabled by new/upgraded diagnostics: high resolution X-Ray crystal spectroscopy; hard X-Ray imaging; active CXRS; bolometry; increased wavenumber spatially localized phase contrast imaging; ultra high speed CCD cameras; IR imaging; swept frequency reflectometry; magnetic pick-ups; and high field side SOL scanning Mach probes.   

Experiments with Lower Hybrid Current Drive on Alcator C-Mod

In 2006 a new 4.6 GHz lower hybrid system was commissioned on Alcator C-Mod, demonstrating the capability of the system in L-mode plasmas. In 2007 the studies have continued on the basic properties of the wave coupling, absorption and current drive as well as those demonstrating the applicability of LHCD to advanced tokamak plasma conditions. In particular, studies of the formation and decay of the fast electron tail have been performed by utilizing fast modulation of the rf power, and coupling studies in H-mode and ICRF dominated plasmas have been performed, yielding conditions under which combined LH and ICRF operation is possible. Operation over a wider range of plasma density and launcher phase has been performed. Injection of LH power in the current ramp phase of the discharge has delayed the onset of the sawtooth oscillation by up to 0.4 s even at modest, $\sim$400 kW, power levels. Measurements with MSE have been made that indicate a broadening of the current profile and comparison of these with the x-ray emission, non-thermal ECE and simulations will be presented.   

Dimensionless pedestal identity plasmas on Alcator C-Mod and JET

Scaling of the H-mode pedestal is crucial for standard scenarios in ITER, but remains uncertain eg owing to the unresolved balance between plasma transport and edge source effects. This can be clarified using plasmas on different tokamaks with identical shape and dimensionless variables at the pedestal top, for which local transport should be the same. ELM-free H-modes at 7.9T, 1.3MA on C-Mod correspond to 1.4T, 0.9MA counterparts on JET, where a power scan in natural-density H-modes with small ELMs matched pedestal identity conditions for higher C-Mod densities. Measurements with a new Thomson scattering diagnostic indicated JET profile widths may be slightly wider than scaled C-Mod values. Modelling of refuelling terms in each device will compare their relative contributions to pedestal structures. Edge magnetic and density fluctuations will also be contrasted. \newline $^{1}$Funded jointly by the UK EPSRC and by the EC under the CoA between EURATOM and UKAEA. \newline $^{2}$Supported by US DOE award DE-FC02-99ER54512.   

H-mode performance and pedestal studies with enhanced particle control on Alcator C-Mod

H-mode density control studies are extended on the Alcator C-Mod tokamak with the recent implementation of an upper chamber cryopump. Experiments have examined the effects of strong neutral pumping on H-mode edge pedestal profiles and core fueling, in varied magnetic topology (both lower and upper null, and close to balanced double null), and over a significant range in edge density. As in prior H-mode puffing experiments, these studies are meant to examine H-mode fueling in discharges with edge neutral opacity approximating that expected in ITER. Significant reduction of edge collisionality is observed with enhanced pumping, concurrent in many cases with high H-mode performance and core density peaking. Cryopumping provides a new tool for obtaining steady H-modes with sustained low collisionality, allowing for continued exploration of pedestal fueling studies, critical-gradient behavior of plasma profiles and access conditions to H-mode regimes (e.g. EDA, ELM-free, ELMy).   

Particle transport and density peaking at low collisionality on Alcator C-Mod

While H-modes tend to have very flat density profiles, modest density peaking is advantageous for fusion performance. Thus robust pinch mechanisms that could allow operation with peaked profiles, in the absence of any internal particle source, are of considerable interest. Recent experiments on C-Mod$^{1}$, at low collisionality, show just such peaking and are quantitatively consistent with earlier results from ASDEX-U$^{2}$ and JET$^{3}$. By extending the range in machine size, these data break the covariance between collisionality and n/n$_{G}$, the density normalized to the density limit and strongly support the primary role of collisionality in determining the profile. The implication is that ITER will have density profiles with $n_e \left( 0 \right)/\left\langle {n_e } \right\rangle \simeq 1.4-1.6$. The C-Mod data also show a small but statistically significant dependence of the peaking factor on the edge safety factor, q. The effect is to increase the peaking by no more than 10{\%} when q is raised from 3.5 to 5.Initial studies of gyrokinetic simulations for these discharges will be shown. $^{1}$M. Greenwald, submitted to Nucl. Fusion, 2007 $^{2}$C. Angioni, et al., PRL 90, 205003, 2003 $^{3}$H.Weisen, Nucl. Fusion 45, L1, 2005   

Ohmic ITBs in Alcator C-Mod

Internal transport barrier plasmas can arise spontaneously in ohmic Alcator C-Mod plasmas where an EDA H-mode has been developed by magnetic field ramping. These ohmic ITBs share the hallmarks of ITBs created with off-axis ICRF injection in that they have highly peaked density and pressure profiles and the peaking can be suppressed by on-axis ICRF. There is a reduction of particle and thermal flux in the barrier region which then allows the neoclassical pinch to peak the central density. Recent work on ITB onset conditions [1] which was motivated by turbulence studies [2] points to the broadening of the T$_{i}$ profile with off-axis ICRF acting to reduce the ion temperature gradient. This suppresses ITG instability driven particle fluxes, which is thought to be the primary mechanism for ITB formation. The object of this study is to examine the characteristics of ohmic ITBs to find whether the stability of plasmas and the plasma parameters support the onset model. \newline [1]K. Zhurovich, et al., To be published in Nuclear Fusion \newline [2] D. R. Ernst, et al., Phys. Plasmas 11, 2637 (2004)   

Hydrogenic Fuel Retention in Molybdenum

High-Z refractory metals such as tungsten and molybdenum (Mo) are favored as plasma-facing components in burning plasma experiment to minimize hydrogenic (H) fuel retention, mainly due to their low H solubility ($\sim $1 appm). Fuel retention in Mo is studied and modeled in the Mo-tile Alcator C-Mod tokamak, and DIONISOS a new facility that features simultaneous plasma bombardment and real-time retention diagnosis. We find that high ion fluxes leads to D trap sites in the Mo; energy wells in which the fuel can reside at concentrations $\sim $ 1{\%}, i.e. much larger than the solubility. The tokamak environment leads to other unique characteristics such as temperature transients through heating and neutron bombardment that further increase retention. High temperature drives D traps and retention deeper into the Mo, but the sudden cooling of the material with removal of the plasma flux ``freezes'' the D deep in the Mo. This physical model recreates the C-Mod retention result, i.e. that a large fraction ($\sim $30{\%}) of the fuelled D can be retained reproducibly over many shots, despite the absence of low-Z film growth. This retention mechanism is fundamentally different than co-deposition of D with carbon, which is observed to dominate D retention in most current tokamaks. The implications for burning plasmas with high neutron loads will be discussed.   

Observations of the Spatial and Temporal Structure of Edge Turbulence near the X-point of Alcator C-Mod

Movies of edge turbulence in the region just outboard of the typical LSN X-point location in C-Mod have been obtained using Gas-Puff-Imaging with a fast-framing (150,000 frames/s) camera. Images of the turbulence in a plane perpendicular to the local field show structures that are highly elongated in the local radial direction. Typically these structures move poloidally, but move only slightly outward in the elongation direction. Both the structure and the motion of the turbulent structures as imaged in this location are very different than the nearly circular cross-sectioned ``blobs'' that are observed near the outer midplane. Field line mapping of circular flux tubes at the midplane show that these are distorted into elongated ``fingers'' when mapped to the viewing location outboard of the X-point, consistent with the observations. Movies of the edge turbulence through L-to-H-mode transitions show no obvious precursor and typically show only a brief ($\sim$1 ms) quiescent period after the transition.   

Interpretation of Edge Turbulence Images near the X-point of Alcator C-Mod

Recent images of edge turbulence taken just outboard of the typical X-point location in Alcator C-Mod show a very different structure and motion from those taken near the outer midplane (see previous talk). We interpret these differences using a model in which the structures are formed near the outer midplane and affect the X-point region through fluctuations propagating along the field lines (most likely electrostatic potential fluctuations). The structural differences are then due to the shearing and flux expansion of the magnetic field between the outer midplane and the X-point, which distort circular `blobs' at the midplane into radial `fingers' near the X-point. Various tests of this model will be discussed and a comparison of the near-X-point images with a BOUT simulation will be described.   

Structure of the Broadband Edge Turbulence in L-mode and pre-H-mode Plasmas in Alcator C-Mod

The edge and near SOL turbulence at the outboard midplane region of ohmic L-mode plasmas have been investigated using gas-puff-imaging and probe measurements. Poloidal wave number information was obtained by a vertical array of views (approximately aligned with a flux-surface) with a 1~MHz sampling rate. We characterized the structure of the broadband turbulence over a range of minor radii that extends both in- and outside the separatrix for various plasma densities, magnetic configurations (LSN, DN, USN) and toroidal fields with a fixed safety factor. The observed dispersion relations show a strong radial structure: the turbulence propagates in the ion- diamagnetic direction at and outside the separatrix, and moves in the electron-diamagnetic direction just inside the separatrix. The dispersions are largely linear with velocities consistent with probe measurements (1.5--2~km/s in the ion, 3.5--4~km/s in the electron direction), yet seem to be inconsistent with a simple sheared flow model. The turbulence structure shows an observable variation in and just prior to L- H transition periods.   

First Measurements of Total Flow Vector in the C-Mod High-Field Side SOL

A new four-electrode scanning Langmuir probe has been installed and operated on the high-field side of Alcator C-Mod. The \textit{WASP} (Wall-Actuated Scanning Probe) has the ability to take data as deep as a few mm inside the LCFS and to measure the total flow vector (parallel and cross-field) as well as fluctuation-induced fluxes using a tungsten-tipped Gundestrup probe. The WASP has verified many of the plasma flow results reported previously using a single-electrode scanning probe [1], as well as adding some important new observations. Near-sonic parallel flows are observed in the far SOL which reverse direction when the topology is reversed. However, their behavior near the separatrix is different. Parallel flows near the separatrix appear to be clamped near zero in LSN but remain high in USN for normal field direction. Cross-field flows show a region of high shear at the radial location of steep pressure gradients, a feature that is also seen on C-Mod's low-field side scanning probes. These and other new results from the WASP will be presented. \newline [1] B. LaBombard et al., Nucl. Fusion \textbf{44} (2004) 1047.   

Plasma flows due to blobs and drifts in Alcator C-Mod scrape-off layer

Long-range near-sonic parallel plasma flows in Alcator C-Mod tokamak scrape-off (SOL) layer are measured and modelled with UEDGE (2D multi-fluid plasma/neutral transport code). As shown, the dominant mechanism driving such flows is fast non- diffusive intermittent cross-field transport (due to blobs) which is modelled as anomalous convection of ion species with strong ballooning-like poloidal asymmetry (HFS/LFS$<$1/10). Here, the properties of asymmetric-transport driven flows, main chamber and divertor recycling, as well as divertor detachment are studied with UEDGE. The conditions for zonal and circular plasma flow patterns in the SOL are analyzed. The impact of ${\bf{\rm{E}}}{\times}{\bf{\rm{B}}}$ and ${\nabla} {\bf{\rm{B}}}$ plasma drifts on non-diffusive transport driven flows is modelled and discussed.   

Fast Electron Driven Alfv\'{e}n Eigenmodes in Alcator C-Mod

Injecting 300 -- 500 kW of Lower Hybrid Current Drive (LHCD) at the very beginning of the discharge, a series of high frequency (200 -- 800 kHz) bursting magnetic fluctuations are observed. There are three sets of modes in ms bursts separated by about 100 kHz with the frequency and time separation of the bursts increasing logarithmically in time. The modes begin about 2.5 ms after the start of the LHCD pulse and persist as long as the plasma is well coupled to the LH grill. The mode amplitudes, measured with pick-up coils on the outboard limiter, reach $\tilde {B}_\theta \quad \approx $ 2 $\times $ 10$^{-5}$ T. The frequency sweeping and amplitude of these modes are similar to those of ICRF fast ion driven Alfv\'{e}n cascades in the current rise[1], but instead are driven by fast electrons and rotate in the electron diamagnetic drift direction. The bursting behavior of the modes is similar to neutral beam driven fishbones[2]. Measurements with a hard x ray imaging array indicate that the fast electron energy reaches up to 80 keV in these very low density plasmas ($\bar {n}_e \quad \le $ 2 $\times $ 10$^{19}$ m$^{-3})$. \newline \newline [1] J A Snipes, et al, \textit{Phys. Plasmas} \textbf{12} (2005) 056102. [2] K McGuire, et al, \textit{Phys. Rev. Lett.} \textbf{50} (1983) 891.   

First Results From The New High Resolution Imaging X-ray Crystal Spectrometer On Alcator C-Mod

In an effort to improve the diagnostic capabilities for measuring plasma rotation on Alcator C-Mod, an imaging x-ray spectrometer has been designed and installed. This instrument utilizes a spherically bent quartz crystal and a set of 2D x-ray detectors to image the entire plasma cross section with a spectral resolving power of approximately 10,000 with vertical spatial resolution of about 1cm. Line emission from highly ionized states of argon and molybdenum are measured at frame rates up to 200Hz. Using spectral tomographic techniques the line integrated spectra can be inverted to determine impurity density, velocity and temperature profiles. An overview of the instrument, analysis and example profiles are presented. Work supported by USDoE Coop. Agree. No. DE-FC02-99ER54512 {\&} DE-AC02-76CH03073.