Search

Deuterium fueling experiments were conducted with the NSTX Liquid Lithium Divertor (LLD). Lithium evaporation recoated the LLD surface to approximate flowing liquid Li to sustain D retention. In the first experiment with the diverted outer strike point on the LLD, the difference between the applied D gas input and the plasma D content reached very high values without disrupting the plasma, as would normally occur in the absence of Li pumping, and there was also little change in plasma D content. In the second experiment, constant fueling was applied, as the LLD temperature was varied to change the surface from solid to liquid. The D retention was relatively constant, and about the same as that for solid Li coatings on graphite, or twice that achieved without Li PFC coatings. Contamination of the LLD surface was also possible due to compound formation and erosion and redeposition from carbon PFCs.

Experimental studies of the novel snowflake divertor concept (D. Ryutov, Phys. Plasmas 14 (2007) 064502) performed in the NSTX and TCV tokamaks are reviewed in this paper. The snowflake divertor enables power sharing between divertor strike points, as well as the divertor plasma-wetted area, effective connection length and divertor volumetric power loss to increase beyond those in the standard divertor, potentially reducing heat flux and plasma temperature at the target. It also enables higher magnetic shear inside the separatrix, potentially affecting pedestal MHD stability. Experimental results from NSTX and TCV confirm the predicted properties of the snowflake divertor. In the NSTX, a large spherical tokamak with a compact divertor and lithium-coated graphite plasma-facing components (PFCs), the snowflake divertor operation led to reduced core and pedestal impurity concentration, as well as reappearance of Type I ELMs that were suppressed in standard divertor H-mode discharges. In the divertor, an otherwise inaccessible partial detachment of the outer strike point with an up to 50% increase in divertor radiation and a peak divertor heat flux reduction from 3-7 MW/m(2) to 0.5-1 MW/m(2) was achieved. Impulsive heat fluxes due to Type-I ELMs were significantly dissipated in the high magnetic flux expansion region. In the TCV, a medium-size tokamak with graphite PFCs, several advantageous snowflake divertor features (cf. the standard divertor) have been demonstrated: an unchanged L-H power threshold, enhanced stability of the peeling-ballooning modes in the pedestal region (and generally an extended second stability region), as well as an H-mode pedestal regime with reduced (x2-3) Type I ELM frequency and slightly increased (20-30%) normalized ELM energy, resulting in a favorable average energy loss comparison to the standard divertor. In the divertor, ELM power partitioning between snowflake divertor strike points was demonstrated. The NSTX and TCV experiments are providing support for the snowflake divertor as a viable solution for the outstanding tokamak plasma-material interface issues. (C) 2013 Elsevier B.V. All rights reserved.

NSTX 2010 experiments were conducted using a molybdenum Liquid Lithium Divertor (LLD) surface installed on the outer part of the lower divertor. This tested the effectiveness of maintaining the deuterium retention properties of a static liquid lithium surface when refreshed by lithium evaporation as an approximation to a flowing liquid lithium surface. The LLD molybdenum front face has a 45% porosity to provide sufficient wetting to spread 37 g of lithium, and to retain it in the presence of magnetic forces. Lithium Evaporators were used to deposit lithium on the LLD surface. At the beginning of discharges, the LLD lithium surface ranged from solid to liquefied depending on the amount of applied and plasma heating. Noteworthy improvements in plasma performance were obtained similar to those obtained previously with lithiated graphite, e.g., ELM-free, quiescent edge, H-modes. During these experiments with the plasma outer strike point on the LLD, the rate of deuterium retention in the LLD, as indicated by the fueling needed to achieve and maintain stable plasma conditions, was the about the same as that for solid lithium coatings on the graphite prior to the installation of the LLD, i.e., about two times that of no-lithium conditions. The role of lithium impurities in this result is discussed. Following the 2010 experimental campaign, inspection of the LLD found mechanical damage to the plate supports, and other hardware resulting from forces following plasma current disruptions. The LLD was removed, upgraded, and reinstalled. A row of molybdenum tiles was installed inboard of the LLD for 2011 experiments with both inner and outer strike points on lithiated molybdenum to allow investigation of lithium plasma facing issues encountered in the first testing of the LLD.

Lithium wall coating techniques have been experimentally explored on National Spherical Torus Experiment (NSTX) for the last six years. The lithium experimentation on NSTX started with a few milligrams of lithium injected into the plasma as pellets and it has evolved to a dual lithium evaporation system which can evaporate up to ∼160 g of lithium onto the lower divertor plates between re-loadings. The unique feature of the NSTX lithium research program is that it can investigate the effects of lithium coated plasma-facing components in H-mode divertor plasmas. This lithium evaporation system has produced many intriguing and potentially important results. In 2010, the NSTX lithium program has focused on the effects of liquid lithium divertor (LLD) surfaces including the divertor heat load, deuterium pumping, impurity control, electron thermal confinement, H-mode pedestal physics, and enhanced plasma performance. To fill the LLD with lithium, 1300 g of lithium was evaporated into the NSTX vacuum vessel during the 2010 operations. The routine use of lithium in 2010 has significantly improved the plasma shot availability resulting in a record number of plasma shots in any given year. In this paper, as a follow-on paper from the 1st lithium symposium [1] , we review the recent progress toward developing fundamental understanding of the NSTX lithium experimental observations as well as the opportunities and associated R&D required for use of lithium in future magnetic fusion facilities including ITER.

NSTX experiments have explored lithium evaporated on a graphite divertor and other plasma-facing components in both L- and H- mode confinement regimes heated by high-power neutral beams. Improvements in plasma performance have followed these lithium depositions, including a reduction and eventual elimination of the HeGDC time between discharges, reduced edge neutral density, reduced plasma density, particularly in the edge and the SOL, increased pedestal electron and ion temperature, improved energy confinement and the suppression of ELMs in the H-mode. However, with improvements in confinement and suppression of ELMs, there was a significant secular increase in the effective ion charge Zeff and the radiated power in H-mode plasmas as a result of increases in the carbon and medium-Z metallic impurities. Lithium itself remained at a very low level in the plasma core

Steady-state handling of divertor heat flux is a critical issue for present and future conventional and spherical tokamaks with compact high power density divertors. A novel “snowflake” divertor (SFD) configuration that takes advantage of magnetic properties of a second-order poloidal null has been predicted to have a larger plasma-wetted area and a larger divertor volume, in comparison with a standard first-order poloidal X-point divertor configuration. The SFD was obtained in 0.8 MA, 4–6 MW NBI-heated H-mode discharges in NSTX using two divertor magnetic coils. The SFD led to a partial detachment of the outer strike point even in low-collisionality scrape-off layer plasma obtained with lithium coatings in NSTX. Significant divertor peak heat flux reduction and impurity screening have been achieved simultaneously with good core confinement and MHD properties.

Major developments on the Mega Amp Spherical Tokamak (MAST) have enabled important advances in support of ITER and the physics basis of a spherical tokamak (ST) based component test facility (CTF), as well as providing new insight into underlying tokamak physics. For example, L–H transition studies benefit from high spatial and temporal resolution measurements of pedestal profile evolution (temperature, density and radial electric field) and in support of pedestal stability studies the edge current density profile has been inferred from motional Stark effect measurements. The influence of the q-profile and E × B flow shear on transport has been studied in MAST and equilibrium flow shear has been included in gyro-kinetic codes, improving comparisons with the experimental data. H-modes exhibit a weaker q and stronger collisionality dependence of heat diffusivity than implied by IPB98(y,2) scaling, which may have important implications for the design of an ST-based CTF. ELM mitigation, an important issue for ITER, has been demonstrated by applying resonant magnetic perturbations (RMPs) using both internal and external coils, but full stabilization of type-I ELMs has not been observed. Modelling shows the importance of including the plasma response to the RMP fields. MAST plasmas with q > 1 and weak central magnetic shear regularly exhibit a long-lived saturated ideal internal mode. Measured plasma braking in the presence of this mode compares well with neo-classical toroidal viscosity theory. In support of basic physics understanding, high resolution Thomson scattering measurements are providing new insight into sawtooth crash dynamics and neo-classical tearing mode critical island widths. Retarding field analyser measurements show elevated ion temperatures in the scrape-off layer of L-mode plasmas and, in the presence of type-I ELMs, ions with energy greater than 500 eV are detected 20 cm outside the separatrix. Disruption mitigation by massive gas injection has reduced divertor heat loads by up to 70%.

Recent experiments on the National Spherical Torus eXperiment (NSTX) have shown the benefits of solid lithium coatings on carbon PFC's to diverted plasma performance, in both L- and H-mode confinement regimes. Better particle control, with decreased inductive flux consumption, and increased electron temperature, ion temperature, energy confinement time, and DD neutron rate were observed. Successive increases in lithium coverage resulted in the complete suppression of ELM activity in H-mode discharges. A liquid lithium divertor (LLD), which will employ the porous molybdenum surface developed for the LTX shell, is being installed on NSTX for the 2010 run period, and will provide comparisons between liquid walls in the Lithium Tokamak eXperiment (LTX) and liquid divertor targets in NSTX. LTX, which recently began operations at the Princeton Plasma Physics Laboratory, is the world's first confinement experiment with full liquid metal plasma-facing components (PFCs). All materials and construction techniques in LTX are compatible with liquid lithium. LTX employs an inner, heated, stainless steel-faced liner or shell, which will be lithium-coated. In order to ensure that lithium adheres to the shell, it is designed to operate at up to 500–600 °C to promote wetting of the stainless by the lithium, providing the first hot wall in a tokamak to operate at reactor-relevant temperatures. The engineering of LTX will be discussed.

Lithium wall coating techniques have been experimentally explored on NSTX for the last five years. The lithium experimentation on NSTX started with a few milligrams of lithium injected into the plasma as pellets and it has evolved to a lithium evaporation system which can evaporate up to ~ 100 g of lithium onto the lower divertor plates between lithium reloadings. The unique feature of the lithium research program on NSTX is that it can investigate the effects of lithium in H-mode divertor plasmas. This lithium evaporation system thus far has produced many intriguing and potentially important results; the latest of these are summarized in a companion paper by H. Kugel. In this paper, we suggest possible implications and applications of the NSTX lithium results on the magnetic fusion research which include electron and global energy confinement improvements, MHD stability enhancement at high beta, ELM control, H-mode power threshold reduction, improvements in radio frequency heating and non-inductive plasma start-up performance, innovative divertor solutions and improved operational efficiency.

NSTX high power divertor plasma experiments have used in succession lithium pellet injection (LPI), evaporated lithium, and injected lithium powder to apply lithium coatings to graphite plasma facing components. In 2005, following the wall conditioning and LPI, discharges exhibited edge density reduction and performance improvements. Since 2006, first one, and now two lithium evaporators have been used routinely to evaporate lithium onto the lower divertor region at total rates of 10-70 mg/min for periods 5-10 min between discharges. Prior to each discharge, the evaporators are withdrawn behind shutters. Significant improvements in the performance of NBI heated divertor discharges resulting from these lithium depositions were observed. These evaporators are now used for more than 80% of NSTX discharges. Initial work with injecting fine lithium powder into the edge of NBI heated deuterium discharges yielded comparable changes in performance. Several operational issues encountered with lithium wall conditions, and the special procedures needed for vessel entry are discussed. The next step in this work is installation of a liquid lithium divertor surface on the outer part of the lower divertor.

Two lithium evaporators were used to evaporate more than 100 g of lithium on to the NSTX lower divertor region. Prior to each discharge, the evaporators were withdrawn behind shutters, where they also remained during the subsequent HeGDC applied for periods up to 9.5 min. After the HeGDC, the shutters were opened and the LITERs were reinserted to deposit lithium on the lower divertor target for 10 min, at rates of 10–70 mg/min, prior to the next discharge. The major improvements in plasma performance from these lithium depositions include: (1) plasma density reduction as a result of lithium deposition; (2) suppression of ELMs; (3) improvement of energy confinement in a low-triangularity shape; (4) improvement in plasma performance for standard, high-triangularity discharges; (5) reduction of the required HeGDC time between discharges; (6) increased pedestal electron and ion temperature; (7) reduced SOL plasma density; and (8) reduced edge neutral density.

Over 100 g of lithium was evaporated onto the NSTX plasma facing components using two stainless steel ovens (LiTER). Plasma emission was recorded using a 16-channel tangential bolometer and a VUV spectrometer. Confinement improved after lithium deposition and ELM’s were either reduced or eliminated for hundreds of milliseconds. The radiated power profiles were generally, but not universally, highly peaked. Both peakness and a modeled on-axis high-Z impurity concentration were used as metrics for the degree of central metallic impurity accumulation. Several changes in plasma performance were seen, including reduction in plasma density and improvements in energy confinement, however high-Z impurity accumulation correlated most directly with ELM-suppression. In the more severe cases of accumulation, the volume-integrated radiated power exceeded 50% of the total input power and the estimated concentration of metals on-axis (modeled using iron as the representative impurity) reached 0.2% of the electron density.

Lithium evaporated onto plasma facing components in the NSTX lower divertor has made dramatic improvements in discharge performance. As lithium accumulated, plasmas previously exhibiting robust Type 1 ELMs gradually transformed into discharges with intermittent ELMs and finally into continuously evolving ELM-free discharges. During this sequence, other discharge parameters changed in a complicated manner. As the ELMs disappeared, energy confinement improved and remarkable changes in edge and scrape-off layer plasma properties were observed. These results demonstrate that active modification of plasma surface interactions can preempt large ELMs.

Coaxial Helicity Injection (CHI) has been successfully used in the National Spherical Torus Experiment (NSTX) for a demonstration of closed flux current generation without the use of the central solenoid. The favorable properties of the Spherical Torus (ST) arise from its very small aspect ratio. However, small aspect ratio devices have very restricted space for a substantial central solenoid. Thus methods for initiating the plasma current without relying on induction from a central solenoid are essential for the viability of the ST concept. CHI is a promising candidate for solenoid-free plasma startup in a ST. The method has now produced closed flux current up to 160 kA verifying the high current capability of this method in a large ST built with conventional tokamak components.

We present the first results of lower divertor heat flux scaling with input power ( P NBI ) and plasma current ( I p ) in H-mode discharges in the National Spherical Torus Experiment. These experiments were conducted with 0.6 ⩽ I p ⩽ 0.9 MA, 1 P NBI B t = 0.45 T, in a lower-single null discharge shape with elongation κ ∼ 2.0, triangularity δ ∼ 0.45, and the ion grad- B drift toward the X -point. As P NBI was raised from 1 MW to 6 MW, the outer divertor peak heat flux was observed to increase up to 6 MW/m 2 . A break in the slope of peak heat flux vs. power was observed at heating power ∼3 MW, which we interpret as the transition of the divertor toward the high recycling regime and well away from the detachment threshold. The peak heat flux was also observed to increase strongly with I p .