Session JO1: C-Mod and JET

Overview of Recent Alcator C-Mod Results

An emphasis of C-Mod research is to address physics and technological issues related to ITER and future burning plasma devices. Among the most critical is plasma facing materials. Recent experiments have indicated that RF sheaths localized to field lines connected to the antenna were responsible for the erosion of the boron layer and subsequent impurity generation on the outer divertor. To obtain broad current profiles, efficient off-axis current drive is required. Initial experiments with LH current drive have observed up to 200 kA of current can be driven for 0.6 MW coupled power. The H-mode transition, pedestal, and dynamics have significant impact on plasma performance. For discharges with high L-$>$H mode threshold, a improvement in the thermal confinement is observed before improved particle confinement. Fast framing camera images of ELMs and intermittent turbulent structures show they travel coherently through the entire SOL. The radial velocity distribution peaks at about 1{\%} of the ion sound speed. Massive gas-jet impurity puffing disruption mitigation has reduced disruption divertor surface heating and halo currents by $\sim $50{\%} at high plasma pressures and short disruption times. The measured evolution of Alfven cascades agrees well with NOVA-K simulations and a number of additional modes have been observed and simulated during sawteeth and EDA H-mode.   

Lower Hybrid Current Drive on Alcator C-MOD

A Lower Hybrid Current Drive (LHCD) system has been installed on the Alcator C-MOD tokomak at MIT. This system was designed for maximum flexibility in the launched parallel wave-number spectrum. This should allow tailoring of the lower hybrid deposition under a variety of plasma conditions. Initial results from this system will be presented. Power levels up to 800 kW have been injected into the tokomak. The parallel wave number has been varied over a wide range, n$_{\vert \vert } \sim $ 2-4. Driven currents up to $\sim $280 kA have been inferred from magnetic measurements, in reasonable agreement with modeling. Sawtooth oscillations vanish and central q is inferred to rise above unity, indicating off-axis CD as expected. Measurements of non-thermal x-ray and electron cyclotron emission confirm the presence of a significant fast electrons population that varies with phase and plasma density. This variation will be compared to that predicted by detailed propagation and absorption codes.   

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, at low collisionality, show just such peaking and are consistent with earlier results from ASDEX-U\footnote{C. Angioni, et al., PRL 90, 205003, 2003.} and JET\footnote{H.Weisen, Nucl. Fusion 45, L1, 2005.} The experiments reported here extend the range of collisionality in C-Mod H-modes downward by almost an order of magnitude. The ratio of central to pedestal density $n_e(\psi = 0)/n_e(\psi = 0.95)$ exceeds 2.0 and $n_e(\psi = 0.4)/n_e(\psi = 0.8)$, which is a measure of peaking in the confinement zone, exceeds 1.5 Wall fueling is exceptionally shallow, in these plasmas, due to their high absolute density ($> 1.5 \times 10^{20}/m^3$) and no beams are used for heating, therefore there is no ambiguity due to competing sources of core particles. Initial gyrokinetic simulations for these discharges will be shown.   

Critical edge gradients and flows with reversed magnetic field in Alcator C-Mod

Both L- and H-mode gradients near the separatrix in C-Mod appear to be set by a ‘critical gradient’ condition [1] -- at fixed values of normalized collisionality, pressure gradient scales as plasma current squared, holding the MHD ballooning parameter, $\alpha_{mhd}$, unchanged. With normal magnetic field direction, lower-null (LSN) L-mode discharges attain higher $\alpha_{mhd}$ values than upper-null (USN). In addition, near-sonic scrape-off-layer flows on the high-field side are co-current directed in LSN and counter in USN, suggesting that such flows may influence critical gradient values. Recently, LSN/USN discharges with reversed magnetic field have been studied, revealing persistent trends: regardless of $B$-field direction, higher $\alpha_{mhd}$ values are attained when $B\times\nabla B$ points towards the x-point -- a generalized condition that results in co-current directed flow in the high-field SOL. The latter observation supports the hypothesis that ballooning-like cross-field transport is responsible for the near-sonic parallel plasma flows observed there [2].\par {\noindent [1] {\it Nucl. Fusion} {\bf 45} (2005) 1658.} \par {\noindent [2] {\it Nucl. Fusion} {\bf 44} (2004) 1047.}   

Simulations of gas jet disruption mitigation

MHD simulations of Alcator C-Mod [V.A Izzo, \textit{Nucl. Fusion }\textbf{46} (2006) 541] have shown that shallow penetration of a high pressure gas jet can lead to a thermal quench of the plasma triggered by growth of an m=2/n=1 instability. Further investigation of this disruption mitigation strategy is carried out through simulations that include a more sophisticated gas jet model and accurate atomic/radiation physics. We seek both to produce a predictive code applicable beyond C-Mod, and to answer specific questions pertaining to the C-Mod experiments, such as the relative importance of impurity mixing and heat transport, the mixing efficiency of low vs. high Z impurities, and the relationship between radiation efficiency, jet penetration speed, and MHD onset. The code deposits neutrals on the grid and tracks each charge state as impurities ionize, recombine, and radiate. Plasma density and temperature are self-consistently evolved, and impurity ions are advected along with the bulk deuterium plasma. Both short simulations at the true C-Mod parameters and longer simulations at reduced Lundquist number are carried out for argon and helium. Preliminary results are presented.   

Using mixed gases for massive gas injection disruption mitigation on Alcator C-Mod

Massive gas injection of noble gases has shown promising results on several tokamaks for mitigating the potential damage cause by disruptions. JT-60U experiments using conventional gas puffing for mitigation previously showed promising results from mixing the gases. We report the use of mixed noble gases with massive gas injection on Alcator C-Mod in order to optimize radiation efficiency, halo current reductions and the overall response time of the mitigation system. Gas-mixtures of helium with incremental fractions of argon (0-50\%) were used. Experiments show that injecting the He/Ar mixtures lead to faster thermal and current quenches as compared to a pure helium or argon injection, thus improving the time response and halo current reduction. Small amounts of argon also lead to optimized radiation fractions with large electron density increases in the core plasma. These results are partially explained by the expectation that small fractions of argon will be entrained with the faster helium at the early phase of the flow. The gas mixing allows one to simultaneously exploit the faster particle delivery rate of the helium with the large radiation capability of the argon.   

Structure and Characteristics of the Quasi-Coherent Mode in EDA H-mode Plasmas

The quasi-coherent mode (QCM), an edge fluctuation present in Enhanced $D_\alpha$ (EDA) H-mode confinement in Alcator C-Mod, is thought to have a decisive role in enhancing particle transport through the edge pedestal of these plasmas. We present detailed data of the mode structure both in real and spectral space, the propagation speed and direction in various regimes, and additional information on the resistive ballooning character of the mode (e.g.~mode propagation with $\mathbf{k}\cdot\mathbf{B}\approx0$ in the counter-current direction). We see a second harmonic at twice the frequency and poloidal wavenumber of the fundamental of the mode ($k_\theta^{fund}$ at $z = 0$ varies between 1-2 cm$^{-1}$); a radial phase variation over the $\sim1$ cm region across the pedestal where the mode is present; and a ballooning-like poloidal variation in amplitude. These observations will be used to examine the strengths and weaknesses of different models of the QC fluctuation, and should be of use in understanding its mechanism.   

Comparison of SOL Turbulence in Limited and Diverted Plasmas in Alcator C-Mod

Edge turbulence in the scrape-off layer (SOL) will be important in determining the power and particle flux on the divertor surfaces in future devices such as ITER. One of the factors determining the SOL turbulence may be the parallel connection to the wall. This relationship has been studied in Alcator C-Mod by comparing the SOL turbulence in diverted plasmas with inner wall limited plasmas under similar conditions. The SOL turbulence was measured near the outer midplane with gas puff imaging and Langmuir probes. Comparisons of the fluctuation levels, frequency spectra, and correlation lengths and times will be made. Initial comparisons will also be made with an edge turbulence simulation using an electromagnetic gyrofluid model.   

Investigation of Energetic ICRF Minority Protons on Alcator C-Mod

Energetic minority protons with $\sim $100keV effective temperature are routinely created in C-Mod plasmas with the application of ICRF. A new multi-channel Compact Neutral Particle Analyzer is used to make measurements of these distributions via an active charge-exchange (CX) technique. Using a detailed model that accounts for beam, halo, and impurity CX, core proton temperatures of $\sim $30-100keV are observed for lower density (n$_{e0}\sim $1-1.5x10$^{20}$/m$^{3})$ C-Mod plasmas using only $\sim $0.5MW of ICRF power. The model found that these fast minority protons are peaked spatially away from r/a=0, even for an on-axis resonance. Additionally, noticeable phase-space anisotropy is seen as expected for ICRF heating. The measured effective temperatures scale approximately with the Stix parameter. Preliminary comparisons with results from the AORSA/CQL3D Full-wave/Fokker-Planck code using a new synthetic diagnostic show good agreement and demonstrate that these complex codes are required to simulate C-Mod's energetic minority populations with accuracy.   

Alfv\'{e}n Eigenmodes in Enhanced D-alpha H-mode in Alcator C-Mod

Unstable Alfv\'{e}n eigenmodes are observed in Enhanced D-alpha (EDA) H-mode in Alcator C-Mod at least up to densities of $\bar {n}_e \sim 2.5\times 10^{20}\mbox{m}^{-3}$ with strong H-minority ICRF heating. The mode frequencies are typically in the range 400 - 800 kHz. The largest amplitude modes ($\tilde {B}_\theta /B_\theta \sim 5\times 10^{-5}$ at the wall) often agree in frequency with the expected Toroidal Alfv\'{e}n Eigenmode (TAE) frequency ($\omega _{TAE} =\mbox{v}_A /2qR)$ for q=1, indicating that these are core modes. The calculated effective fast ion tail energy $\sim $ 50 keV corresponds to $\mbox{v}_F /\mbox{v}_A \sim 0.4$. These modes are observed to rotate in the electron diamagnetic drift direction, which is opposite to the rotation direction expected for TAEs with a peaked fast ion distribution. Recent calculations with the AORSA and CQL3D codes [1] indicate that the fast ion distribution in C-Mod may be hollow because of poor wave focussing and preferential heating of trapped particles along the resonance. The opposite sign of the gradient of a hollow profile of the fast ion $\beta $ would then explain the opposite direction of rotation. NOVA-K modelling will be used to compare the stability of L and H-mode conditions and determine the radial structure of these modes. \newline [1] E. F. Jaeger, et al, to be published in Plasma Physics and Cont. Fus.   

Study of triggering mechanisms for internal transport barriers in Alcator C-Mod

Internal transport barriers (ITBs) can be routinely produced in C-Mod steady enhanced D$_{\alpha }$ (EDA) H-mode plasmas by applying ICRF at $\vert $r/a$\vert \quad \ge $ 0.5 (off-axis heating). Access to the off-axis ICRF heated ITBs may be understood within the paradigm of marginal stability. Analysis of the electron temperature profiles shows a decrease of a/L$_{Te}$ in the ITB region as the RF resonance is moved off axis with no apparent difference for the very core and edge plasma regions. Ion temperature profiles are calculated using the transport analysis code TRANSP. These calculations also reveal a decrease of a/L$_{Ti}$ in the ITB region as the ICRF resonance is moved outward. Furthermore, there is experimental evidence that Ti profiles broaden as ICRF changes from on-axis to off-axis heating. Stability analysis is performed using the gyrokinetic code GS2. Linear GS2 calculations do not reveal any difference in ETG growth rate profiles for ITB vs. non-ITB discharges. However, they show that the region of stability to ITG modes gets wider as the ICRF is moved outward. Supported by USDoE award DE-FC02-99ER54512   

Cross-field transport asymmetries in unbalanced double-null divertor configurations.

A key feature of anomalous cross-field transport in the tokamak SOL is strong poloidal asymmetry where the plasma flux is much larger on low field side (LFS) than on high field side (HFS). As shown, this asymmetry can cause an enhanced main chamber recycling at LFS and the large, M$\sim $1, parallel plasma flows. Recent experiments performed in Alcator C-Mod with an unbalanced double null (DN) configuration show (as measured by reciprocating probes at the LFS and HFS midplanes) that plasma density profiles have a much shorter cross-field decay length in the region outside the secondary separatrix on the HFS compared to the same region on the LFS. In well-balanced DN discharges, the density e-folding length measured at the HFS midplane is substantially smaller than that in single null (SN) configuration. These data independently indicate a strong ballooning-like asymmetry. We use UEDGE code to simulate plasma transport in unbalanced DN shots obtained in C-mod. From matching experimental profile data, we infer an asymmetry factors for the anomalous cross-field plasma transport (diffusive and convective) which are indicative of strong asymmetry. The effect of secondary separatrix on radial plasma profiles, parallel plasma flows, and impurity migration will be discussed. Simulated transport asymmetries in unbalanced DN, balanced DN, and SN shots will be compared. Work supported by DoE grant DEFG0204ER54739.   

Development of Benign Elm Scenarios at JET

The development of a Q=10 baseline regime with tolerable ELMs remains a challenging issue for ITER. To address this, the JET programme has focused on the exploration of intrinsically benign ELM plasma scenarios. At high shape and fuelling type I ELM frequencies are strongly reduced by inter-ELM turbulence. Similar turbulence can lead to a fully stationary pedestal at high collisionality in near double-null (ASDEX Upgrade identity) shapes. At higher \textit{$\beta $}$_{p}$ and $q_{95}$ very small ELMs ("Grassy") are seen. JET has now been upgraded with a new divertor to allow more highly shaped plasmas. Preliminary results in the new ITER-like shape show a strong sensitivity of ELM type and confinement to X point location. Further studies using the new divertor to access quasi double-null configurations with increased lower triangularity and decreased upper triangularity have produced the first fully stationary type-II-like regimes with good confinement. Work is now under way to resolve the detailed access requirements and behaviour of benign ELM regimes, testing in particular the role of double-null proximity, collisionality, \textit{$\beta $}$_{p}$ and $q_{95}$. Full results will be reported at conference.   

Qualifying The Hybrid Scenario for ITER at JET

The hybrid scenario is an attractive alternative regime for ITER allowing long pulse operation with little or no loss of performance. New simulations from JET indicate fusion gains $>$10 and pulses lengths of 1000-3000s in ITER. One of the key questions is whether the hybrid scenario is a new scenario in a confinement sense or if the main benefit is an improved MHD stability allowing for operation at higher beta than the baseline scenario. Analysis with Weiland and GLF23 models indicate little difference in core transport processes -- H$_{IPB98y2}$ factors of $\sim $1.2 in JET might be an artefact of a too-negative \textit{$\beta $} dependence in the scaling law. Controlled comparisons between the regimes are underway to test this and explore scaling with \textit{$\beta $}. Most hybrid regime discharges are at low collisionality where other studies have identified density peaking effects. This may raise the performance but may also cause impurity accumulation. New modelling shows that this can be ameliorated by electron heating driving TEM instabilities. Studies will test this thesis and also explore extension of the parameter space to low $q_{95}$, high density, high \textit{$\beta $}, and the physics of long pulse operation.   

Resolving Tearing Physics for the ITER Baseline at JET

New results from JET are elucidating the physics and exploring control of tearing instabilities for the ITER baseline scenario. In particular use of lower hybrid current drive at low magnetic fields (where deposition is expected nearer the edge) demonstrates an alternate option for mode control. At high beta this led to $\sim $15-50{\%} reductions in 3/2 neoclassical tearing mode (NTM) amplitudes. It also improved resilience of otherwise-unheated plasmas to error field driven 2/1 tearing modes, raising error field thresholds by $\sim $20{\%}. For ITER a further concern is posed by its naturally low rotation. New JET results show a trend of 3/2 NTMs being triggered at lower beta as neutral beam momentum injection is reduced. The most serious concern arises from 2/1 modes locking to cause disruptions. However use of the newly installed Error Field Correction Coils shows that with suitable error field correction, this can be avoided - thus good error field correction will be needed in the low rotation plasmas of ITER. These results are part of an integrated programme to optimise the stability of the baseline scenario at JET, to be reported in full at conference.