Search

A major research goal of the national spherical torus experiment is establishing long-pulse, high beta, high confinement operation and its physics basis. This research has been enabled by facility capabilities developed during 2001 and 2002, including neutral beam (up to 7?MW) and high harmonic fast wave (HHFW) heating (up to 6?MW), toroidal fields up to 6?kG, plasma currents up to 1.5?MA, flexible shape control, and wall preparation techniques. These capabilities have enabled the generation of plasmas with of up to 35%. Normalized beta values often exceed the no-wall limit, and studies suggest that passive wall mode stabilization enables this for H mode plasmas with broad pressure profiles. The viability of long, high bootstrap current fraction operations has been established for ELMing H mode plasmas with toroidal beta values in excess of 15% and sustained for several current relaxation times. Improvements in wall conditioning and fuelling are likely contributing to a reduction in H mode power thresholds. Electron thermal conduction is the dominant thermal loss channel in auxiliary heated plasmas examined thus far. HHFW effectively heats electrons, and its acceleration of fast beam ions has been observed. Evidence for HHFW current drive is obtained by comparision of the loop voltage evolution in plasmas with matched density and temperature profiles but varying phases of launched HHFW waves. Studies of emissions from electron Bernstein waves indicate a density scale length dependence of their transmission across the upper hybrid resonance near the plasma edge that is consistent with theoretical predictions. A peak heat flux to the divertor targets of 10?MW?m?2 has been measured in the H mode, with large asymmetries being observed in the power deposition between the inner and outer strike points. Non-inductive plasma startup studies have focused on coaxial helicity injection. With this technique, toroidal currents up to 400?kA have been driven, and studies to assess flux closure and coupling to other current drive techniques have begun.

Research on the spherical torus (or spherical tokamak) (ST) is being pursued to explore the scientific benefits of modifying the field line structure from that in more moderate aspect ratio devices, such as the conventional tokamak. The ST experiments are being conducted in various US research facilities including the MA-class National Spherical Torus Experiment (NSTX) at Princeton, and three medium sized ST research facilities: PEGASUS at University of Wisconsin, HIT-II at University of Washington, and CDX-U at Princeton. In the context of the fusion energy development path being formulated in the US, an ST-based Component Test Facility (CTF) and, ultimately a Demo device, are being discussed. For these, it is essential to develop high performance, steady-state operational scenarios. The relevant scientific issues are energy confinement, MHD stability at high beta (?), non-inductive sustainment, Ohmic-solenoid-free start-up, and power and particle handling. In the confinement area, the NSTX experiments have shown that the confinement can be up to 50% better than the ITER-98-pby2 H-mode scaling, consistent with the requirements for an ST-based CTF and Demo. In NSTX, CTF-relevant average toroidal beta values ?T of up to 35% with a near unity central ?T have been obtained. NSTX will be exploring advanced regimes where ?T up to 40% can be sustained through active stabilization of resistive wall modes. To date, the most successful technique for non-inductive sustainment in NSTX is the high beta poloidal regime, where discharges with a high non-inductive fraction (~60% bootstrap current+NBI current drive) were sustained over the resistive skin time. Research on radio-frequency (RF) based heating and current drive utilizing high harmonic fast wave and electron Bernstein wave is also pursued on NSTX, PEGASUS, and CDX-U. For non-inductive start-up, the coaxial helicity injection, developed in HIT/HIT-II, has been adopted on NSTX to test the method up to Ip ~ 500?kA. In parallel, start-up using a RF current drive and only external poloidal field coils are being developed on NSTX. The area of power and particle handling is expected to be challenging because of the higher power density expected in the ST relative to that in conventional aspect-ratio tokamaks. Due to its promise for power and particle handling, liquid lithium is being studied in CDX-U as a potential plasma-facing surface for a fusion reactor.

Key DIII-D advanced tokamak (AT) experimental and modelling results are applied to examine the physics and control issues for ITER to operate in a negative central shear (NCS) AT scenario. The effects of a finite edge pressure pedestal and current density are included based on the DIII-D experimental profiles. Ideal and resistive stability analyses demonstrate that feedback control of resistive wall modes by rotational drive or flux conserving intelligent coils is crucial for these AT configurations to operate at attractive βN values in the range 3.0–3.5. Vertical stability and halo current analyses show that reliable disruption mitigation is essential and mitigation control using an impurity gas can significantly reduce the local mechanical stress to an acceptable level. Core transport and turbulence analyses indicate that control of the rotational shear profile is essential to reduce the pedestal temperature required for high β. Consideration of edge stability and core transport suggests that a sufficiently wide pedestal is necessary for the projected fusion performance. Heat flux analyses indicate that, with core-only radiation enhancement, the outboard peak divertor heat load is near the design limit of 10 MW m−2. Detached operation may be necessary to reduce the heat flux to a more manageable level. Evaluation of the ITER pulse length using a local step response approach indicates that the 3000 s ITER long-pulse scenario is probably both necessary and sufficient for demonstration of local current profile control.

Recent experiments on the DIII-D tokamak have focused on determining the effect of trapped particles on the electron cyclotron current drive (ECCD) efficiency. The experimental ECCD efficiency increases as the deposition location is moved towards the inboard midplane or towards smaller minor radius for both co- and counter-injection; the ECCD efficiency also increases with increasing electron density and/or temperature. The experimental ECCD is compared to both the linear theory (TORAY-GA) as well as a quasilinear Fokker–Planck model (CQL3D) and is found to be in better agreement with the more complete Fokker–Planck calculation, especially when the rf power density and/or loop voltage exceed criterion for substantial nonlinear modification of the electron distribution function. The width of the measured ECCD profile is consistent with the theoretically expected width in the absence of radial transport for the current carrying electrons.

Recent experiments on DIII-D have demonstrated the ability to sustain plasma conditions that integrate and sustain the key ingredients of advanced tokamak (AT) operation: high β with 1.5 2.0, plasmas with β≈2.9% and 90% of the plasma current driven non-inductively have been sustained for nearly 2 s (limited only by the duration of the ECCD pulse). Negative central magnetic shear is produced by the ECCD, leading to the formation of a weak internal transport barrier even in the presence of Type I ELMs. Separate experiments have demonstrated the ability to sustain a steady current density profile using ECCD for periods as long as 1 s with β = 3.3% and >90% of the current driven non-inductively. In addition, stable operation well above the ideal no-wall β limit has been sustained for several energy confinement times with the duration only limited by resistive relaxation of the current profile to an unstable state. Stability analysis indicates that the experimental β limit depends on the degree to which the no-wall limit can be exceeded and weakly on the actual no-wall limit. Achieving the necessary density levels required for adequate ECCD efficiency requires active divertor exhaust and reducing the wall inventory buildup prior to the high performance phase. Simulation studies indicate that the successful integration of high β operation with current profile control consistent with these experimental results should result in high β, fully non-inductive plasma operation.

The prevention of neoclassical tearing modes (NTMs) in tokamak plasmas is a major challenge for fusion. Ideal modes can seed NTMs through forced reconnection, yet in sawtoothing discharges it is not well understood why a particular sawtooth crash seeds a NTM after several preceding sawteeth did not. Also, tearing modes sometimes appear and grow without an obvious ideal mode causing a seed island. Based on theoretical and experimental results a new mechanism for tearing mode onset is proposed and tested which explains these puzzling observations. Tearing stability calculations based on experimental equilibria from the DIII-D tokamak [J. L. Luxon and L. G. Davis, Fusion Technol. 8, 441 (1985)] indicate that tearing modes can be driven unstable by a rapid increase in the linear tearing stability index Δ′ just before onset. Near the ideal kink limit in βN≡β/(I/aBT), Δ′ becomes large and positive as it approaches a pole discontinuity. The relative time scales of the change in Δ′ vs the effects of finite island width will determine the evolution of the island, and the eventual nonlinear state. Theoretical predictions of the onset point, the βN at a specified small island width, and the early evolution characteristics are compared with results from new DIII-D experiments designed specifically to test this hypothesis. Time integration of the island evolution equation shows that these predictions are consistent with the experimental observations. The nonlinear 3D resistive magnetohydrodynamic code NIMROD [A. H. Glasser et al., Plasma Phys. Control. Fusion 41, A747 (1999)] is used to evolve equilibria reconstructed from these experiments in time to provide a more comprehensive prediction of the island evolution during the early nonlinear phase. The simulations are also consistent with the proposed mechanism.

Significant progress has been made on the HT-7 superconducting tokamak (R=1.22 m, a=0.27 m) in the past year toward obtaining advanced operating modes in steady state. By combining ion Bernstein wave and lower hybrid current drive edge-localized-mode-free limiter H-mode discharges with H89∼2, βN*H89>3 have been obtained that lasted for 53 τE. A well-boronized wall prevented further impurity accumulation and the large pumping kept the recycling at a very low level. The density was kept almost constant with the feedback control. The noninductive driven current fraction is about 60% and bootstrap fraction is about 18%. The central electron temperature is about 2.6 keV and density above 2×1019 m−3. With reduced plasma parameters of central Te∼1.5 keV and ne0∼2×1019 m−3, H89∼1.45, βN∼1.1, the improved confinement phase lasts for more than 100 τE with the noninductive driven current fraction ∼70%.

Using off-axis electron cyclotron current drive (ECCD), self-consistent integrated advanced tokamak operation has been demonstrated on DIII-D [Plasma Physics and Controlled Nuclear Fusion Research, 1986 (International Atomic Energy Agency, Vienna, 1987), Vol. I, p. 159], combining high β (>3%) at high q (qmin>2.0) with good energy confinement (H89∼2.5) and high noninductive current fraction (fBS∼55%,fNI∼90%). Modification of the current profile by ECCD led to internal transport barrier formation even in the presence of type I edge localized modes. Improvements were observed in all transport channels, and increased peaking of profiles led to higher bootstrap current in the core. Separate experiments have shown the ability to maintain a nearly steady-state current profile for up to 1 s with qmin>1.5. Modeling indicates that this favorable current profile can be maintained indefinitely at a higher βN using tools available to the near-term DIII-D program. Modeling and simulation have become essential tools fo...

Recently a number of major, unanticipated effects have been reported in tokamak edge research raising the question of whether we understand the controlling physics of the edge. This report is on the first part – here focused on the outer divertor – of a systematic study of the simplest possible edge plasma – no ELMs, no detachment, etc. – for a set of 10 repeat, highly diagnosed, single-null, divertor discharges in DIII-D. For almost the entire, extensive data set so far evaluated, the matches of experiment and model are so close as to imply that the controlling processes at the outer divertor for these simple plasma conditions have probably been correctly identified and quantitatively characterized in the model. The principal anomaly flagged so far relates to measurements of Te near the target, potentially pointing to a deficiency in our understanding of sheath physics in the tokamak environment.

This paper describes the modelling of the feedback control and rotational stabilization of the resistive wall mode (RWM) in tokamaks. A normal mode theory for the feedback stabilization of the RWM has been developed for an ideal plasma with no toroidal rotation. This theory has been numerically implemented for general tokamak geometry and applied to the DIII-D tokamak. A general formulation is further developed for the feedback stabilization of a tokamak with toroidal rotation and plasma dissipation. It has been used to understand the role of the external resonant field in affecting the plasma stability and compared with the resonant field amplification phenomenon observed in DIII-D. The effectiveness of a differentially rotating resistive wall in stabilizing the RWM has also been studied numerically. It is found that for a non-circular tokamak, a wide range of flow patterns are all effective. The structure of the RWM predicted from ideal MHD theory has been compared with signals from various diagnostics. It is also projected that based on DIII-D results scaled up to the ITER-FEAT, 33 MW of 1 MeV negative neutral beam injection will be able to sustain plasma rotation sufficient to stabilize the RWM without relying on feedback.

The National Spherical Torus Experiment (NSTX) at Princeton University, Princeton, NJ, is designed to assess the potential of the low-aspect-ratio spherical torus concept for magnetic plasma confinement. The plasma has been heated by up to 7 MW of neutral beam injection (NBI) at an injection energy of 100 keV and up to 6 MW of high harmonic fast wave (HHFW) at 30 MHz. NSTX has achieved /spl beta//sub T/ of 32%. A variety of MHD phenomena have been observed to limit /spl beta/. NSTX has now begun addressing /spl tau//sub E/ scaling, /spl beta/ limits, and current drive issues. During the NBI heating experiments, a broad T/sub i/ profile with T/sub i/ up to 2 keV, T/sub i/>T/sub e/ and a large toroidal rotation were observed. Transport analysis suggests that the impurity ions have diffusivities approaching neoclassical. For L-Mode plasmas, /spl tau//sub E/ is up to two times the ITER97L L-Mode scaling and exceeds the ITER98pby2 H-Mode scaling in some cases. Transitions to H-Mode have been observed which result in an approximate doubling of /spl tau//sub E/ after the transition in some conditions. During HHFW heating, T/sub e/>T/sub i/ and T/sub e/ up to 3.5 keV were observed. Current drive has been studied using both coaxial helicity injection with up to 390 kA of toroidal current and HHFW. HHFW has produced H-modes with significant bootstrap current fraction at low I/sub p/, high q, and high /spl beta//sub p/.

A magnetohydrodynamic (MHD) instability which has the characteristics that the displacement phase is in the same direction at all the affected flux surfaces (an interchange-like structure with no phase reversal across a flux surface), has been observed during the L mode in several negative central shear discharges of DIII-D tokamak [J. L. Luxon, P. Anderson, F. Baity et al., Plasma Physics and Controlled Fusion Research 1986 (International Atomic Energy Agency, Vienna, 1987), Vol. I, p. 159]. The instability occurs when the minimum safety factor is around 2.0 and the profile of the safety factor q is deeply reversed in the center. Detailed stability analyses were carried out using standard numerical codes and the high quality magnetic probe, motional Stark effect and electron cyclotron emission (ECE) data. Analysis of the data after the onset shows that the instability has an n=1 mode number and a growth time of about 400 μs. The electron temperature fluctuations obtained from ECE measurements indicate a ...

Electron cyclotron current drive (ECCD) experiments on the DIII-D tokamak are solidifying the physics basis for localized, off-axis current drive, the goal being to validate a predictive model for ECCD. The ECCD profiles are determined from the magnetic field pitch angles measured by motional Stark effect polarimetry. The measured ECCD switches from the co- to the counter-direction as the toroidal injection angle is varied with a profile width that is in accordance with ray tracing calculations. Tests of electron trapping in low beta plasmas show that the ECCD efficiency decreases rapidly as the deposition is moved off-axis and towards the outboard side of the plasma, but the detrimental effects of electron trapping on the current drive are greatly reduced in high beta plasmas. Overall, the measured ECCD is in good agreement with theoretical calculations using a quasilinear Fokker–Planck code over a wide range of injection angles and plasma parameters.

The tearing mode stability of low β plasmas is studied experimentally in the DIII-D [J. L. Luxon, Nucl. Fusion 42, 614 (2002)] tokamak. The linear and nonlinear characteristics of the plasma are measured and compared with theoretical predictions. In contrast to the neoclassical tearing mode, which occurs at high β and is dominated by the bootstrap current effect, in the low β regime the neoclassical bootstrap current effect is minimal. Thus the stability properties are more amenable to comparison with presently available theoretical codes. In the linear phase, the onset of the instability has been found to be predictable from equilibrium reconstruction to agree with the availability of the tearing mode free energy. In the nonlinear phase, the temperature and its fluctuation amplitude and phase have been fitted to the prediction of achieving a perturbed three-dimensional equilibrium with an assumed perturbation eigenfunction.

Results from an array of theoretical and computational tools developed to treat the instabilities of most interest for high performance tokamak discharges are described. The theory and experimental diagnostic capabilities have now been developed to the point where detailed predictions can be productively tested so that competing effects can be isolated and either eliminated or confirmed. The linear MHD stability predictions using high quality discharge equilibrium reconstructions are tested against the observations for the principal limiting phenomena in DIII-D: L mode negative central shear (NCS) disruptions, H mode NCS edge instabilities, and tearing and resistive wall modes (RWMs) in long pulse discharges. In the case of predominantly ideal plasma MHD instabilities, agreement between the code predictions and experimentally observed stability limits and thresholds can now be obtained to within several per cent, and the predicted fluctuations and growth rates to within the estimated experimental errors. Edge instabilities can be explained by a new model for edge localized modes as predominantly ideal instabilities with low to intermediate toroidal mode number. Accurate ideal calculations are critical to demonstrating RWM stabilization by plasma rotation, and the ideal eigenfunctions provide a good representation of the RWM structure when the plasma rotation slows. Ideal eigenfunctions can then be used to predict stabilization using active feedback. For non-ideal modes, the agreement in some cases is promising. Δ' calculations, for example, indicate that some discharges are linearly unstable to classical tearing modes, consistent with the observed growth of islands in those discharges. Nevertheless, there is still a great deal of improvement required before the non-ideal predictive capability can routinely approach levels similar to those for the ideal comparisons.

Some analyses that survey experimental data are carried out at a sparse sample rate between pulses during tokamak operation and/or completed as a batch job overnight because the complete analysis on a single fast workstation cannot fit in the narrow time window between two pulses. Scientists therefore miss the opportunity to use these results to guide experiments quickly. With a dedicated Beowulf type cluster at a cost less than that of a workstation, these analyses can be accomplished between pulses and the analyzed data made available for the research team during the tokamak operation. A Linux PC cluster comprised of 12 processors was installed at DIII-D National Fusion Facility in CY00 and expanded to 24 processors in CY01 to automatically perform between-pulse magnetic equilibrium reconstructions using the efit code written in fortran , cer analyses using cerquick code written in interactive data language ( idl) and full profile fitting analyses (ne, Te, Ti, Vr, Zeff) using idl code zipfit . This paper reports the current status of the system, the details of the between-pulse profile fitting analyses, and discusses some problems and concerns raised during the implementation and expansion of the system.

Research on the stability of spherical torus plasmas at and above the no-wall beta limit is being addressed on the National Spherical Torus Experiment [M. Ono et al., Nucl. Fusion 40, 557 (2000)], that has produced low aspect ratio plasmas, R/a∼1.27 at plasma current exceeding 1.4 MA with high energy confinement (TauE/TauE_ITER89P>2). Toroidal and normalized beta have exceeded 25% and 4.3, respectively, in q∼7 plasmas. The beta limit is observed to increase and then saturate with increasing li. The stability factor βN/li has reached 6, limited by sudden beta collapses. Increased pressure peaking leads to a decrease in βN. Ideal stability analysis of equilibria reconstructed with EFIT [L. L. Lao et al., Nucl. Fusion 25, 1611 (1985)] shows that the plasmas are at the no-wall beta limit for the n=1 kink/ballooning mode. Low aspect ratio and high edge q theoretically alter the plasma stability and mode structure compared to standard tokamak configurations. Below the no-wall limit, stability calculations show the perturbed radial field is maximized near the center column and mode stability is not highly effected by a nearby conducting wall due to the short poloidal wavelength in this region. In contrast, as beta reaches and exceeds the no-wall limit, the mode becomes strongly ballooning with long poloidal wavelength at large major radius and is highly wall stabilized. In this way, wall stabilization is more effective at higher beta in low aspect ratio geometry. The resistive wall mode has been observed in plasmas exceeding the ideal no-wall beta limit and leads to rapid toroidal rotation damping across the plasma core.

Sustained stabilization of the n=1 kink mode by plasma rotation at beta approaching twice the stability limit calculated without a wall has been achieved in DIII-D by a combination of error field reduction and sufficient rotation drive. Previous experiments have transiently exceeded the no-wall beta limit. However, demonstration of sustained rotational stabilization has remained elusive because the rotation has been found to decay whenever the plasma is wall stabilized. Recent theory [Boozer, Phys. Rev. Lett. 86, 5059 (2001)] predicts a resonant response to error fields in a plasma approaching marginal stability to a low-n kink mode. Enhancement of magnetic nonaxisymmetry in the plasma leads to strong damping of the toroidal rotation, precisely in the high-beta regime where it is needed for stabilization. This resonant response, or “error field amplification” is demonstrated in DIII-D experiments: applied n=1 radial fields cause enhanced plasma response and strong rotation damping at beta above the no wal...

H-mode operation is the choice for next-step tokamak devices based either on conventional or advanced tokamak physics. This choice, however, comes at a significant cost for both the conventional and advanced tokamaks because of the effects of edge-localized modes (ELMs). ELMs can produce significant erosion in the divertor and can affect the β limit and reduced core transport regions needed for advanced tokamak operation. Recent experimental results from DIII-D have demonstrated a new operating regime, the quiescent H-mode regime, which solves these problems. We have achieved quiescent H-mode operation which is ELM-free and yet has good density control. In addition, we have demonstrated that an internal transport barrier can be produced and maintained inside the H-mode edge barrier for long periods of time (>3.5 s or >25 energy confinement times τE). By forming the core barrier and then stepping up the input power, we have achieved βNH89 = 7 for up to 10 times the τE of 160 ms. The βNH89 values of 7 substantially exceed the value of 4 routinely achieved in standard ELMing \mbox{H-mode.} The key factors in creating the quiescent H-mode operation are neutral beam injection in the direction opposite to the plasma current (counter injection) plus cryopumping to reduce the density. Density control in the quiescent H-mode is possible because of the presence of an edge MHD oscillation, the edge harmonic oscillation, which enhances the edge particle transport while leaving the energy transport unaffected.

The dependence of the edge pedestal electron pressure on magnetic geometry is studied using the international pedestal database. For type I ELM discharges, the pressure normalized to B 2 is found to scale roughly inversely with major radius and inverse aspect ratio. In addition, a strong increase in pressure with increasing shape is found when the pressure is normalized to a ballooning type scaling. Ballooning stability calculations suggest that there is significant variation in the normalized edge pressure gradient in the database. The observed shape dependence reflects this variation, although the applicability of the log-linear type dependences used here must be in question if phase changes such as from first to second stability are spanned by the data. This question can only be resolved by upgrading the database to include edge pressure profiles, or fits to profiles, rather than only pedestal-top values.

Ideal MHD stability limits of anticipated plasma configurations for NSTX [Ono M. et al 2000 Nucl. Fusion 40 557] and the dependence on the parameters defining the MHD equilibrium are evaluated. The study provides a quantitative computational evaluation of the stability limit variations induced by changes to the equilibrium of NSTX high β plasmas. The analysis is based on a reference free-boundary equilibrium with β = 41.5%, monotonic safety factor q profile (qa = 12.1, q0 = 2.8) and broad pressure profile p (peaking factor Fp≡p(0)/p = 1.7). On this reference target local variation of the plasma boundary, and the safety factor q and pressure profiles p are imposed. Localized inflection of the outboard plasma boundary, produced by near field effects from poloidal shaping field coils, weaken the stability due to the destabilization of high n ballooning modes. Variation of the q profile at different radial locations can also degrade stability. Both experimental profiles from existing tokamaks and spherical torus machines and profiles generated from transport modelling of anticipated neutral beam heated plasmas are used. Degraded stability is found at increasing pressure peaking factor due to the destabilization of n = 1 kink/ballooning modes. Direct access to the second region of stability is found in certain configurations and, for the entire set of variations considered, the lower calculated β limit values are still in the range of 20.0% without considering the stabilizing effect of the passive conducting structures.

Substantial progress has been made towards both understanding and control of internal transport barriers (ITBs) on DIII-D, resulting in the discovery of a new sustained high performance operating mode termed the quiescent double barrier (QDB) regime. The QDB regime combines core transport barriers with a quiescent ELM-free H mode edge (termed QH mode), giving rise to separate (double) core and edge transport barriers. The core and edge barriers are mutually compatible and do not merge, resulting in broad core profiles with an edge pedestal. The QH mode edge is characterized by ELM-free behaviour with continuous multiharmonic MHD activity in the pedestal region and has provided density and radiated power control for longer than 3.5 s (25τE) with divertor pumping. QDB plasmas are long pulse high performance candidates, having maintained a βN H89 product of 7 for five energy confinement times (Ti ≤ 16 keV, βN ≤ 2.9, H89 ≤ 2.4 τE ≤ 150 ms, DD neutron rate Sn ≤ 4 × 1015 s-1). The QDB regime has only been obtained in counter-NBI discharges (injection antiparallel to the plasma current) with divertor pumping. Other results include successful expansion of the ITB radius using (separately) both impurity injection and counter-NBI, and the formation of ITBs in the electron thermal channel using both ECH and strong negative central shear (NCS) at high power. These results are interpreted within a theoretical framework in which turbulence suppression is the key to ITB formation and control, and a decrease in core turbulence is observed in all cases of ITB formation.

Experimental and modeling investigations on the DIII-D and EAST tokamaks show the attractive transport and stability properties of fully noninductive, high poloidal-beta (β P ) plasmas, and their suitability for steady-state operating scenarios in ITER and CFETR. A key feature of the high-β P regime is the large-radius (ρ > 0.6) internal transport barrier (ITB), often observed in all channels (ne, Te, Ti, rotation), and responsible for both excellent energy confinement quality and excellent stability properties. Experiments on DIII-D have shown that, with a large-radius ITB, very high β N and β P values (both ≥ 4) can be reached by taking advantage of the stabilizing effect of a nearby conducting wall. Synergistically, higher plasma pressure provides turbulence suppression by Shafranov shift, leading to ITB sustainment independent of the plasma rotation. Experiments on EAST have been used to assess the long pulse potential of the high-β P regime. Using RF-only heating and current drive, EAST achieved minute-long fully noninductive steady state H-mode operation with strike points on an ITER-like tungsten divertor. Improved confinement (relative to standard H-mode) and steady state ITB features are observed with a monotonic q-profile with q min ~ 1.5. Separately, experiments have shown that increasing the density in plasmas driven by lower hybrid wave broadens the q-profile, a technique that could enable a large radius ITB. These experimental results have been used to validate MHD, current drive, and turbulent transport models, and to project the high-β P regime to a burning plasma. These projections suggest the Shafranov shift alone will not suffice to provide improved confinement (over standard H-mode) without rotation and rotation shear. However, increasing the negative magnetic shear (higher q on axis) provides a similar turbulence suppression mechanism to Shafranov shift, and can help devices such as ITER and CFETR achieve their steady-state fusion goals.

Investigation on the equilibrium operation regime, its ideal magnetohydrodynamics (MHD) stability and edge localized modes (ELM) characteristics is performed for the China Fusion Engineering Test Reactor (CFETR). The CFETR operation regime study starts with a baseline scenario (R = 5.7 m, B T = 5 T) derived from multi-code integrated modeling, with key parameters varied to build a systematic database. These parameters, under profile and pedestal constraints, provide the foundation for the engineering design. The long wavelength low-n global ideal MHD stability of the CFETR baseline scenario, including the wall stabilization effect, is evaluated by GATO. It is found that the low-n core modes are stable with a wall at r/a = 1.2. An investigation of intermediate wavelength ideal MHD modes (peeling ballooning modes) is also carried out by multi-code benchmarking, including GATO, ELITE, BOUT++ and NIMROD. A good agreement is achieved in predicting edge-localized instabilities. Nonlinear behavior of ELMs for the baseline scenario is simulated using BOUT++. A mix of grassy and type I ELMs is identified. When the size and magnetic field of CFETR are increased (R = 6.6 m, B T = 6 T), collisionality correspondingly increases and the instability is expected to shift to grassy ELMs.

Transport modeling of a proposed ITER steady-state scenario based on DIII-D high poloidal-beta () discharges finds that ITB formation can occur with either sufficient rotation or a negative central shear q-profile. The high scenario is characterized by a large bootstrap current fraction (80%) which reduces the demands on the external current drive, and a large radius internal transport barrier which is associated with excellent normalized confinement. Modeling predictions of the electron transport in the high scenario improve as approaches levels similar to typical existing models of ITER steady-state and the ion transport is turbulence dominated. Typical temperature and density profiles from the non-inductive high scenario on DIII-D are scaled according to 0D modeling predictions of the requirements for achieving a steady-state fusion gain in ITER with 'day one' heating and current drive capabilities. Then, TGLF turbulence modeling is carried out under systematic variations of the toroidal rotation and the core q-profile. A high bootstrap fraction, high scenario is found to be near an ITB formation threshold, and either strong negative central magnetic shear or rotation in a high bootstrap fraction are found to successfully provide the turbulence suppression required to achieve .

An efficient technique for building plasma equilibria with preset or constrained q profiles by solving the nonlinear Grad–Shafranov equation iteratively is proposed. The technique uses a set of dynamic virtual poloidal field probes in the plasma core region. Numerical examples are shown for monotonic q profiles and negative central magnetic shear equilibria which are obtained with this technique. As a proof of principle, the technique is applied on the EAST tokamak, to constrain the plasma equilibrium reconstruction with the experimental q = 1 surface information from the ECE diagnostic in addition to the external magnetic measurements.

Using the SOLPS5.0 code, an assessment has been undertaken of the relative importance of two key aspects of divertor baffle geometry: (i) divertor closure, and (ii) field-target angle. The figures of merit (FOMs) have been taken to be achieving specified (low) values of electron temperature at the target, T et, and parallel power flux density at the target, q ||t, at the lowest possible 'upstream' density, n e,sep,OMP, for a given power input. As expected, it was found that increased closure for both orthogonal targets and small glancing-angle targets make for a better divertor, based on these FOMs. It was also found that the combination of these aspects is significantly more effective than either aspect separately. The latter appears to be related to a strong correlation between T et and deuterium gas density at the target, which is the subject of a companion paper (Stangeby P.C. and Sang C.F. 2017 Nucl. Fusion 57 056007).

SOLPS-EIRENE edge code analysis shows that a gas-tight slot divertor geometry with a small-angle (glancing-incidence) target, named the small angle slot (SAS) divertor, can achieve cold, dissipative/detached divertor conditions at relatively low values of plasma density at the outside midplane separatrix. SAS exhibits the following key features: (1) strong enhancement of the buildup of neutral density in a localized region near the plasma strike point on the divertor target; (2) spreading of the cooling front across the divertor target with the slot gradually flaring out from the strike point, thus effectively reducing both heat flux and erosion on the entire divertor target surface. Such a divertor may potentially provide a power and particle handling solution for long pulse advanced tokamaks.

A new diagnostic has been developed on DIII-D that determines the impurity poloidal rotation from the poloidal asymmetry in the toroidal angular rotation velocity. This asymmetry is measured with recently added tangential charge exchange viewchords on the high-field side of the tokamak midplane. Measurements are made on co- and counter-current neutral beams, allowing the charge exchange cross section effect to be measured and eliminating the need for atomic physics calculations. The diagnostic implementation on DIII-D restricts the measurement range to the core (r/a < 0.6) where, relative to measurements made with the vertical charge exchange system, the spatial resolution is improved. Significant physics results have been obtained with this new diagnostic; for example, poloidal rotation measurements that significantly exceed neoclassical predictions.

Energy and pitch angle resolved measurements of escaping neutral beam ions (E approximate to 80 keV) have been made during DIII-D L-mode discharges with applied, slowly rotating, n = 2 magnetic perturbations. Data from separate scintillator detectors (FILDs) near and well below the plasma midplane show fast-ion losses correlated with the internal coil (I-coil) fields. The dominant fast-ion loss signals are observed to decay within one poloidal transit time after beam turn-off indicating they are primarily prompt loss orbits. Also, during application of the rotating I-coil fields, outboard midplane edge density and bremsstrahlung emission profiles exhibit a radial displacement of up to delta R approximate to 1 cm. Beam deposition and full orbit modeling of these losses using M3D-C1 calculations of the perturbed kinetic profiles and fields reproduce many features of the measured losses. In particular, the predicted phase of the modulated loss signal with respect to the I-coil currents is in close agreement with FILD measurements as is the relative amplitudes of the modulated losses for the co and counter-current beam used in the experiment. These simulations show modifications to the beam ion birth profile and subsequent prompt loss due to changes in the edge density; however, the dominant factor causing modulation of the losses to the fast-ion loss detectors is the perturbed magnetic field (delta B/B approximate to 10(-3) in the plasma). Calculations indicate total prompt loss to the DIII-D wall can increase with application of the n = 2 perturbation by up to 7% for co-current injected beams and 3% for counter-current injected beams depending on phase of the perturbation relative to the injected beam. US Department of Energy DE-FC02-04ER54698, SC-G903402, DEAC02- 09CH11466, DE-FG02-04ER54761, DE-FG02- 05ER54808

Edge conditions in DIII-D are being quantified in order to provide insight into the physics of the H?mode regime. Several studies show that electron temperature is not the key parameter that controls the L-H transition. Gradients of edge temperature and pressure are much more promising candidates for elements of such parameters. They systematically increase during the L phases of discharges which make a transition to H?mode, and these increases are typically larger than the increases in the underlying quantities. The quality of H?mode confinement is strongly correlated with the height of the H?mode pedestal for the pressure. The gradient of the pressure is limited by MHD modes, in particular by ideal kink ballooning modes with finite mode number n. For a wide variety of discharges, the width of the barrier for electron pressure is well described by a relationship that is proportional to (?pedp)1/2. A new regime of confinement, called the quiescent H?mode, which provides steady state operation with no ELMs, low radiated power and normal H?mode confinement, has been discovered. A coherent edge MHD mode provides adequate particle transport to control the plasma density while permitting the pressure pedestal to remain almost identical to that observed in ELMing discharges.

Experiments on the DIII-D tokamak have identified a new sustained high-performance operating mode, termed the quiescent double barrier (QDB) regime. The QDB regime combines internal transport barriers (ITBs) with a quiescent, edge localized mode (ELM)-free H-mode edge, termed QH-mode, giving rise to separate core and edge transport barriers. These double barriers have been maintained for {>}3.5 s (~25τE), demonstrating a long-pulse, quasi-steady-state capability. The combination of core ITBs and edge H-mode temperature pedestals results in high-performance plasmas; a βN H89 product of 7 has been maintained for 10 τE, other peak (non-simultaneous) parameters include Ti≤17 keV, βN≤2.9% m T MA-1, H89≤2.6, β≤3.8%, τE≤ 160 ms, and DD neutron rate Sn≤5.5×1015 s-1. These results address a major issue with tokamak plasmas: how to sustain long-pulse, high-performance H-mode plasmas without ELMs, yet retaining the density and impurity control hitherto provided by ELMs. In these QDB plasmas ELMs are replaced by continuous benign MHD activity in the edge, which enhances particle transport. A signature of operation with a QH-mode edge appears to be very large radial electric fields in the edge and scrape-off layer (SOL). In the core, simulations and modelling replicate many of the features of the observed transport and fluctuation behaviour, including the ion temperature profile and turbulence correlation lengths. Slow high-Z impurity accumulation (τ≥500 ms) is observed in the centre of many QDB plasmas, and is the subject of ongoing analysis. To date the QDB regime has only been obtained in plasmas with counter-NBI (injection anti-parallel to the plasma current), and with divertor cryopumping to control the density.

Significant progress in obtaining high performance discharges lasting many energy confinement times in the DIII-D tokamak has been realized in recent experimental campaigns. Normalized performance ∼10 has been sustained for more than 5>E with qmin > 1.5. (The normalized performance is measured by the product βNH89, indicating the proximity to the conventional β limits and energy confinement quality, respectively.) These H mode discharges have an ELMing edge and β < 5%. The limit to increasing β is a resistive wall mode, rather than the tearing modes as previously observed. Confinement remains good despite qmin > 1. The global parameters were chosen to optimize the potential for fully non-inductive current sustainment at high performance, which is a key program goal for the DIII-D facility. Measurement of the current density and loop voltage profiles indicate that ≈75% of the current in the present discharges is sustained non-inductively. The remaining ohmic current is localized near the half-radius. The electron cyclotron heating system is being upgraded to replace this remaining current with ECCD. Density and β control, which are essential for operating advanced tokamak discharges, were demonstrated in ELMing H mode discharges with βNH89 ≈ 7 for up to 6.3 s or ≈34τE. These discharges appear to have stationary current profiles with qmin ≈ 1.05 in agreement with the current profile relaxation time ≈1.8 s.

Research in NSTX has been conducted to establish spherical torus plasmas to be used for high ?, auxiliary heated experiments. This device has a major radius R0 = 0.86?m and a midplane halfwidth of 0.7?m. It has been operated with toroidal magnetic field B0 ? 0.3?T and Ip ? 1.0?MA. The evolution of the plasma equilibrium is analysed between discharges with an automated version of the EFIT code. Limiter, double null and lower single null diverted configurations have been sustained for several energy confinement times. The plasma stored energy reached 92?kJ (?t = 17.8%) with neutral beam heating. A plasma elongation in the range 1.6 ? ? ? 2.0 and a triangularity in the range 0.25 ? ? ? 0.45 have been sustained, with values of ? = 2.6 and ? = 0.6 being reached transiently. The reconstructed magnetic signals are fitted to the corresponding measured values with low errors. Aspects of the plasma boundary, pressure and safety factor profiles are supported by measurements from non-magnetic diagnostics. Plasma densities have reached 0.8 and 1.2 times the Greenwald limit in deuterium and helium plasmas, respectively, with no clear limit encountered. Instabilities including sawteeth and reconnection events, characterized by Mirnov oscillations, and a perturbation of the Ip, ? and li evolutions, have been observed. A low q limit was observed and is imposed by a low toroidal mode number kink instability.

The main aim of the National Spherical Torus Experiment (NSTX) is to establish the fusion physics principles of the spherical torus (ST) concept. The NSTX device began plasma operations in February 1999 and the plasma current Ip was successfully brought up to the design value of 1 MA on 14 December 1999. The planned plasma shaping parameters, elongation κ = 1.6-2.2 and triangularity δ = 0.2-0.4, were achieved in inner wall limited, and single null and double null diverted configurations. The coaxial helicity injection (CHI) and high harmonic fast wave (HHFW) experiments were also initiated. CHI current of 27 kA produced up to 260 kA toroidal current without using an ohmic solenoid. With the injection of 2.3 MW of HHFW power, using 12 antennas connected to six transmitters, electrons were heated from a central temperature of 400 eV to 900 eV at a central density of 3.5 × 1013 cm-3, increasing the plasma energy to 59 kJ and the toroidal β, βT, to 10%. The NBI system commenced operation in September 2000. The initial results with two ion sources (PNBI = 2.8 MW) show good heating, producing a total plasma stored energy of 90 kJ corresponding to βT ≈ 18% at a plasma current of 1.1 MA.

Coaxial helicity injection (CHI) on the National Spherical Torus Experiment (NSTX) has produced 240 kA of toroidal current without the use of the central solenoid. Values of the current multiplication ratio (CHI produced toroidal current/injector current) up to 10 were obtained, in agreement with predictions. The discharges, which lasted for up to 200 ms, limited only by the programmed waveform, are more than an order of magnitude longer in duration than any CHI discharges previously produced in a spheromak or a spherical torus.

A 12-processor Linux PC cluster has been installed to perform between-pulse magnetic equilibrium reconstructions during tokamak operations using the EFIT code written in FORTRAN. The MPICH package implementing message passing interface is employed by EFIT for data distribution and communication. The new system calculates equilibria eight times faster than the previous system yielding a complete equilibrium time history on a 25 ms time scale 4 min after the pulse ends. A graphical interface is provided for users to control the time resolution and the type of EFITs. The next analysis to benefit from the cluster is CERQUICK written in IDL for ion temperature profile analysis. The plan is to expand the cluster so that a full profile analysis (Te, Ti, ne, Vr, Zeff) can be made available between pulses, which lays the ground work for Kinetic EFIT and/or ONETWO power balance analyses.

A new sustained high-performance regime, combining discrete edge and core transport barriers, has been discovered in the DIII-D tokamak. Edge localized modes (ELMs) are replaced by a steady oscillation that increases edge particle transport, thereby allowing particle control with no ELM-induced pulsed divertor heat load. The core barrier resembles those usually seen with a low (L) mode edge, without the degradation often associated with ELMs. The barriers are separated by a narrow region of high transport associated with a zero crossing in the E x B shearing rate.

Significant progress has been made in obtaining high-performance discharges for many energy confinement times in the DIII-D tokamak [J. L. Luxon et al., Plasma Physics and Controlled Fusion Research (International Atomic Energy Agency, Vienna, 1987), Vol. I, p. 159]. Normalized performance (measured by the product of βNH89 and indicative of the proximity to both conventional β limits and energy confinement quality, respectively) ∼10 has been sustained for >5 τE with qmin>1.5. These edge localized modes (ELMing) H-mode discharges have β∼5%, which is limited by the onset of resistive wall modes slightly above the ideal no-wall n=1 limit, with approximately 75% of the current driven noninductively. The remaining Ohmic current is localized near the half-radius. The DIII-D electron cyclotron heating system is being upgraded to replace this inductively driven current with localized electron cyclotron current drive (ECCD). Density control, which is required for effective ECCD, has been successfully demonstrated in long-pulse high-performance ELMing H-mode discharges with βNH89∼7 for up to 6.3 s. In plasma shapes compatible with good density control in the present divertor configuration, the achieved βN is somewhat less than that in the high βNH89=10 discharges.

The mission of the National Spherical Torus Experiment (NSTX) is to extend the understanding of toroidal physics to low aspect ratio (R/a approximately equal to 1.25) in low collisionality regimes. NSTX is designed to operate with up to 6 MW of High Harmonic Fast Wave (HHFW) heating and current drive, 5 MW of Neutral Beam Injection (NBI) and Co-Axial Helicity Injection (CHI) for non-inductive startup. Initial experiments focused on establishing conditions that will allow NSTX to achieve its aims of simultaneous high-bt and high-bootstrap current fraction, and to develop methods for non-inductive operation, which will be necessary for Spherical Torus power plants. Ohmic discharges with plasma currents up to 1 MA and with a range of shapes and configurations were produced. Density limits in deuterium and helium reached 80% and 120% of the Greenwald limit respectively. Significant electron heating was observed with up to 2.3 MW of HHFW. Up to 270 kA of toroidal current for up to 200 msec was produced noninductively using CHI. Initial NBI experiments were carried out with up to two beam sources (3.2 MW). Plasmas with stored energies of up to 140 kJ and bt =21% were produced.

External impurity injection into L mode edge discharges in DIII-D has produced clear confinement improvement (a factor of 2 in energy confinement and neutron emission), reduction in all transport channels (particularly ion thermal diffusivity to the neoclassical level), and simultaneous reduction of long wavelength turbulence. Suppression of the long wavelength turbulence and transport reduction are attributed to synergistic effects of impurity induced enhancement of E × B shearing rate and reduction of toroidal drift wave turbulence growth rate. A prompt reduction of density fluctuations and local transport at the beginning of impurity injection appears to result from an increased gradient of toroidal rotation enhancing the E × B shearing. Transport simulations carried out using the National Transport Code Collaboration demonstration code with a gyro-Landau fluid model, GLF23, indicate that E × B shearing suppression is the dominant transport suppression mechanism.

The dependence of edge stability on plasma shape and local pressure gradients P' in the DIII-D and JT-60U tokamaks is studied. The stronger plasma shaping in DIII-D allows the edge region of DIII-D discharges with type I (giant) ELMs to have access to the second region of stability for ideal ballooning modes and a larger edge pedestal pressure gradient P' than JT-60U type I ELM discharges. These JT-60U discharges are near the ballooning mode first regime stability limit. The DIII-D results support an ideal stability based working model of type I ELMs as low to intermediate toroidal mode number, n, MHD modes. The results from a stability analysis of JT-60U type I ELM discharges indicate that the predictions of this model are also consistent with JT-60U edge stability observations.

The first equilibrium reconstruction of EAST current-density profile based on internal Faraday rotation measurements provided by the POlarimeter-INTerferometer (POINT) diagnostic is demonstrated using the EFIT equilibrium reconstruction code. EFIT incorporates 11 simultaneous line-integrated density and Faraday effect measurements from POINT to self-consistently reconstruct the equilibrium toroidal current density profile using a Faraday rotation reconstruction algorithm. It is shown that the POINT measurements can be applied to improve the accuracy of core plasma current density and q profile on EAST. Comparisons of magnetic surfaces and the q profile reconstructed using external magnetic data against those using magnetic and POINT data are presented. Equilibrium reconstructions using POINT data are found to be consistent with sawtooth phenomena. The sensitivity of equilibrium reconstruction to POINT measurements indicates Faraday rotation provides important constraints for determining the current profile.

Scrape-off layer plasma simulation modeling has been carried out to assess the effect of tightly closing the lower divertor in DIII-D, which at present is almost fully open, on the achievement of cold dissipative/detached divertor conditions. To isolate the impact of other factors on the divertor plasma solution and to make direct comparisons, most of the parameters including the meshes were kept as similar as possible. Only the neutral baffling was modified to compare a fully open divertor with a tightly closed one. The modeling shows that the tightly closed divertor greatly improves trapping of recycling neutrals, thereby increasing radiative and charge exchange losses in the divertor and reducing the electron temperature T et and deposited power density q dep at the target plate. Furthermore, the closed structure enables the divertor plasma to enter into highly dissipative and detached divertor conditions at a significantly lower upstream density. The effects of divertor closure on the neutral density and pressure, and their correlation with the divertor plasma conditions are also demonstrated. The effect of molecular D2−ion D+ elastic collisions and neutral–neutral collisions on the divertor plasma solution are assessed.

The frequency and amplitude of edge localized instabilities in DIII-D tokamak H mode discharges are shown to depend on the discharge shape squareness. An abrupt increase in instability frequency and decreases in instability amplitude and edge pressure gradient are observed when the squareness is changed to a sufficiently large or a sufficiently small value. The observed changes in the instability frequency and amplitude correlate with a predicted loss, at high and low values of discharge shape squareness, of access in the H mode edge region to the ideal infinite toroidal mode number ballooning mode second regime of stability. When no edge region second stability regime access is predicted, the measured pressure gradient is in agreement with the predicted ballooning mode first stability regime threshold. When the second stability regime is predicted to be locally accessible, the edge region pressure gradient increases to values well above the first stability regime limit.

The status of modelling work focused on developing the advanced tokamak (AT) scenarios in DIII-D is discussed. The objective of the work is twofold: (a) to develop AT scenarios with ECCD using time dependent transport simulations, coupled with heating and current drive models, consistent with MHD equilibrium and stability; and (b) to use time dependent simulations to help plan experiments and to understand the key physics involved. Time dependent simulations based on transport coefficients derived from experimentally achieved target discharges are used to perform AT scenario modelling. The modelling indicates that off-axis ECCD with approximately 3 MW absorbed power can maintain high performance discharges with qmin > 1 for 5-10 s. The resultant equilibria are calculated to be stable to n = 1 pressure driven modes. The plasma is well into the second stability regime for high-n ballooning modes over a large part of the plasma volume. The role of continuous localized ECCD is studied for stabilizing m/n = 2/1 tearing modes. Progress towards validating current drive and transport models, consistent with experimental results, and developing self-consistent, integrated high performance AT scenarios is discussed.

Recent progress in the development of long pulse, high performance discharges on the DIII-D tokamak is reviewed. This is highlighted by a discharge achieving simultaneously a ?N H of 9, bootstrap current fraction of 0.5 and non-inductive current fraction of 0.75, sustained for 16 energy confinement times. The physics challenge has changed in the long pulse regime. Non-ideal MHD modes limit the stability, fast ion driven modes may play a role in fast ion transport, which limits the stored energy, and plasma edge behaviour can affect the global performance. New control tools are being developed to address these issues.

In JT-60U H-mode plasmas, giant (type I) ELMs disappear and minute grassy ELMs appear when triangularity δ, edge safety factor q95 and βp are high enough. Complete suppression of giant ELMs was observed at δ0.45, q956 and βp1.6. At higher δ (0.54), giant ELMs can disappear at a lower q95 (~4.0). In the grassy ELMy H-mode, edge temperature and pressure can be higher than those in giant ELMy H-mode and a favourable confinement can be sustained without an increase of the impurity concentration. An edge stability analysis suggests that the edge plasma is accessing the second stability regime of the high n ballooning mode in the grassy ELMy discharges.

The amplitude and frequency of modes driven in the edge region of tokamak high mode (H-mode) discharges [type I edge-localized modes (ELMs)] are shown to depend on the discharge shape. The measured pressure gradient threshold for instability and its scaling with discharge shape are compared with predictions from ideal magnetohydrodynamic theory for low toroidal mode number (n) instabilities driven by pressure gradient and current density and good agreement is found. Reductions in mode amplitude are observed in discharge shapes with either high squareness or low triangularity where the stability threshold in the edge pressure gradient is predicted to be reduced and the most unstable mode is expected to have higher values of n. The importance of access to the ballooning mode second stability regime is demonstrated through the changes in the ELM character that occur when second regime access is not available. An edge stability model is presented that predicts that there is a threshold value of n for second r...

The characteristics of the H-mode are studied in discharges with varying triangularity and squareness. The pressure at the top of the H-mode pedestal increases strongly with triangularity, primarily due to an increase in the margin by which the edge pressure gradient exceeds the ideal ballooning mode first stability limit. Two models are considered for how the edge may exceed the ballooning mode limit. In one model, access to the ballooning mode second stable regime allows the edge pressure gradient and associated bootstrap current to continue to increase until an edge localized, low toroidal mode number, ideal kink mode is destabilized. In the second model, the finite width of the H-mode transport barrier and diamagnetic effects raise the pressure gradient limit above the ballooning mode limit. We observe a weak inverse dependence of the width of the H-mode transport barrier, Δ, on triangularity relative to the previously obtained scaling Δ∝(βPEP)1/2. The energy loss for Type I ELMs increases with triangularity in proportion to the pedestal energy increase. At low density, the energy confinement of high-triangularity discharges is larger than discharges with low triangularity, as a result of an increase in the energy in the H-mode pedestal. At high density, both the change in pedestal pressure and the response of the density profile are found to play a role in setting the energy confinement. The highest energy confinement at high density was obtained in low-triangularity discharges.