Pellet Injection Experiments Research Articles

Pellet ablation and mass deposition in FTU: analysis of vertical and low field side injection experiments

Pellet injection is considered the most promising technique to fuel ITER plasmas. However, standard low field side (LFS) pellet injection seems inadequate for an efficient fuelling of the plasma core because it is too demanding in terms of injection velocity. Alternative configurations, as injection from the high field side (HFS) or vertically from the top of the machine (VHFS) are technologically less demanding due to a drift of the ablated material down the magnetic field gradient, which helps its penetration. In the past year LFS and VHFS pellet injection experiments have been performed on the Frascati Tokamak Upgrade (FTU: BT = 8 T, Ip = 1.5 MA, R = 93.5 cm, a = 30 cm), a circular cross-section device with a molybdenum toroidal limiter located on the HFS. The rationale of these experiments was to study the advantages of VHFS injection in a high toroidal field machine, where the value of BT can be considered as representative of that in ITER. The results demonstrate that the main advantage of VHFS injection (with respect to LFS) is that it allows one to obtain the same post-pellet density profile with an injection speed 2–3 times lower than that required in the case of LFS injection. However no significant displacement of the ablated material was observed, possibly due to the small value of the ∇ B-induced drift in these ohmic, low β plasmas. The experimental measurements were compared to the predictions of the drift model (Pégourié B. et al 2007 Nucl. Fusion 47 44) showing a reasonable agreement in both the magnitude of the drift and the main features of the line integrated density during pellet ablution.

Spectroscopic Measurement Using Wide Range UV/Visible Spectroscopic System GAMMA 10

Impurity spectra have been measured to evaluate impurity behavior and plasma parameter in the tandem mirror GAMMA 10. An ultraviolet and visible (UV/visible) spectroscopic system has been designed to measure the impurity emission intensity in detail. It consists of two spectrometers to obtain an entire wavelength range of UV/visible impurity spectra in one plasma shot. The time-varying emission intensity of radiation spectra have been measured successfully with electron cyclotron resonance heating (ECRH) or pellet injection plasmas. We evaluate radiation loss with ECRH from the GAMMA 10 plasma in the UV/visible range; further we estimate the electron density and temperature after applying the spectral intensity data measured from the pellet injection experiment to a collisional-radiative model.

High-Speed Visible Imaging of the Central-Cell Plasmas in the GAMMA 10 Tandem Mirror

Details of two-dimensional imaging performed in the central-cell of the GAMMA 10 tandem mirror by using high-speed camera are described. At the GAMMA 10 central-cell midplane, a high-speed camera (MEMRECAM fx-K4, NAC Inc.) was mounted and the shape and the motion of plasmas were precisely investigated. In the standard ion cyclotron range of frequency (ICRF) heated plasmas, gas puffing of hydrogen with short pulse (3-5 ms) close to the central-cell midplane was carried out to illuminate the periphery plasma and the time evolution of visible light emitted from the gas was captured. The Hα line-emission was localized close to the port of gas puffer. In the pellet injection experiments, 2-D images of the ablation light induced by hydrogen ice-pellet injection were successfully captured at the GAMMA 10 central-cell. Analysis of 2-D images clarified that the electron cyclotron heating applied in the central-cell (c-ECH) has a significant influence on the edge plasma turbulence and the ablation of hydrogen ice-pellet.

Electron Density Measurement by Using a Multi-Channel Interferometer System in the Tandem Mirror GAMMA 10

Fluctuation in the plasma is important to be measured for studying the improvement of the plasma confinement by the formation of the plasma confinement potential. Density fluctuation is observed using microwaves, such as interferometer, reflectometry and Fraunhofer diffraction method. We have constructed a new multi-channel microwave interferometer for measuring the plasma density and fluctuation radial profiles in a single plasma shot. In order to study higher density plasma operation, we have started sub-millimeter hydrogen ice pellet injection experiments into the potential confined plasma. We successfully measured the time dependent density and line-integrated density fluctuation radial profiles in the pellet injection experiments using the multi-channel microwave interferometer.

Repetitive pellet fuelling for high-density/steady-state operation on LHD

A repetitive pellet injector has been developed and pellet refuelling experiments have been launched on the Large Helical Device (LHD) in a new regime characterized by high density steady state operation. In order to suppress the disturbance to the core plasma as much as possible, we employ relatively small and slow pellets. The size and velocity of the pellets are typically 2.5 mm in diameter and 250 m s−1, respectively. The pellets were injected into neutral beam heated plasmas with repetition frequency of 10 Hz. The density rise per pellet is 1.5 × 1019 m−3 and the pellet penetration depth, which is normalized by the plasma minor radius, is typically less than 0.5. Quasi-steady-state operation was achieved at plasma parameters of , Te = Ti = 1.0 keV. Repetitive pellet fuelling showed improved energy confinement compared with massive gas puff fuelling in the high density regime in common with previous transient pellet injection experiments, and it can sustain such an improved confinement property.

Performance of fueling pellet injectors for the large helical device

Techniques of solid hydrogen pellet injection into plasmas towards steady state operation as well as further reliable hardware have been demonstrated for the large helical device (LHD). The conversion of hardware is performed to attain higher reliability and reproducibility of pellets. Major issues to improve the quality of pellets are the cooling temperature of pellet production by a cryo-cooler and the incident angle to the inner wall of the guide tube for pellets. The former is improved from 12.6 to 10.6 K by the reduction of the heat conduction and radiation to produce the robust pellets. The latter is optimized from 5.41° to 1.73° at the maximum by the low-angle pellet path to mitigate the destruction of pellets in the guide tube. It has been confirmed by these conversion that the quality of pellets is significantly improved, i.e., the probability of intact pellets is increased to ∼100%. Also, the controllability of the pellet size is desired as needs of pellet injection experiments. It has been demonstrated by the property of pellet production related to the cooling temperature and the pressure of supplied hydrogen gas.

Fuelling and Diagnostics with Pellet Injection in the HL-1M Tokamak

The pellet injection experiments for fuelling and diagnostics have been carried out on the HL-1M tokamak. The eight-pellet injector was installed on HL-1M. A reliable monitor-detector and camera system was set up to take initial pellet photographs and measure the initial pellet speed and size. High fuelling efficiency of 60%-100% and a density profile with a peaking factor of 1.8-2.0 were obtained. The maximum density close to 1014/cm3 in HL-1M was achieved with newly optimized combined fuelling techniques. Two typical models of pellet ablation have been utilized for simulative calculation of the ablation rates in HL-1M. In comparison with the distribution of the measured Hα emission intensity from the digital data of the CCD camera, the experimental result seems more optimistic for core fuelling than theoretical predictions by the two models. The safety factor profile q(r) has been extracted from the information provided by the CCD camera during the pellet injection. The reliability of the measured results depends mainly on the calibration of the imaging space position. Based on the calibration, the measured q-profile becomes more reasonable than those published previously for the same shot number and same photograph.

Gas and Pellet Injection Systems for JT-60 and JT-60U

Designs and operations of the gas system and pellet injection systems for JT-60 and JT-60U are described. A gas injection valve that is a key component of the gas injection system was developed using a multilayer piezoelectric element. The maximum flow rate of this system is 43.3 Pa⋅m3/s. The valve has mechanism for adjustment at atmospheric side meaning that a repair and an adjustment can be conducted without ventilation inside a vacuum vessel. It was confirmed that the effect of magnetic field and temperature change on the valves in the JT-60U environment was negligible.In JT-60U, two systems of pellet injector – a pneumatic drive and a centrifugal one – were developed. The pneumatic type attained a pellet velocity of 2.3 km/s, which was the world record at the time in 1988. On the other hand, the centrifugal one was developed in 1998. This injector can eject trains of up to 40 cubic (2.1 mm3) pellets at frequencies of 1 to 10 Hz and speed of 0.1 to 1.0 km/s. A guide tube for a magnetic high field side injection (HFS) (top) was also developed in 1999. The pellet injection experiment with the HFS system started in 2000. In addition, another guide tube for HFS(mid) injection was newly developed and installed in March 2001. These systems are working well.

Preliminary Observation on Coordination of Pellet Injectionand Ion Bernstein Wave on a HT-7 Tokamak

A pellet injection (PI) experiment was performed during the application ofthe ion Bernstein wave on a HT-7 tokamak. A preliminary coordination effect wasobserved. With a lower wave power, shortly after PI, the coupling of the wave wasenhanced, and the particle confinement was improved. With higher power,off-axis heating for 15% at about a/3 in the low field side was observed.

Analysis and modelling of plasma response to pellet injection in RFX

Pellet injection experiments have been performed on the RFX reversed field pinch (RFP). The main aim of the experimental campaign was the study of the target plasma response to get information both on the possibility of an efficient pellet fuelling of the discharge and on the associated transport modification. In this paper, we present an analysis of different experimental scenarios. The tool employed is a one-dimensional transport code coupled with an ablation calculation that yields the pellet particle source and trajectory. It is seen that, in standard plasmas, the pellets that reach the plasma centre trigger a phase of rapid profile flattening and poorer particle and energy confinement, and an increase of the transport coefficients is required to reproduce the post-pellet behaviour of the profiles. On the other hand, the slow pellets that suffer large deflection and miss the plasma centre do not have such a deleterious effect and, in this case, the unperturbed transport coefficients are able to reproduce the post-pellet evolution of the profiles. Finally, when the pellets are launched in combination with some method that induces an improved confinement, the transport degradation is less evident. In our case, pellet injection has been combined with pulsed poloidal current drive (PPCD), a technique that reduces magnetic field stochasticization and transport in the RFP core. When the effect of the PPCD is strong enough, a synergy between pellet injection and PPCD is observed and plasma fuelling without dramatic confinement reduction is possible.

A repetitive pellet injection system for JT-60U

A centrifugal pellet injector and a guide tube for magnetic high-field side (HFS) injection were developed in JT-60U. The injector can eject trains of up to 40 cubic (1.9 mm) 3 deuterium pellets at frequencies of 1–10 Hz and velocities of 0.2–1.0 km/s. Pellet injection experiments into ohmic heating (OH) plasma were performed. From plasma density measurement, central fueling and enhanced fueling rate compared to gas puffing were observed for magnetic low-field side (LFS) injections. The results of HFS injection through the guide tube connected to the upper port of the vessel showed that the pellet velocity exceeding 220 m/s caused pellet destruction inside the tube.

Control System of Pellet Injector on the HT-7 Tokamak

In this paper the control system of the pellet injector is introduced in detail and the system mainly includes two parts: The present and the remote control system. The present control system controls the injector and provides the interface to the remote system. And the remote control system has acquired present signals with analog input card and perform the actions through digit output card, it also has an interface for Windows programming easily used by the operators when carrying out the pellet injection experiments. Through several HT-7 campaigns, the remote control system has been validated to be feasible and reliable and has made successful shots for studying the interactions between the pellets and plasma.

Analysis of pellet ablation and density build-up in the RFXreversed-field pinch

New data from the Reversed Field eXperiment (RFX) are presented and analysed,which provide a deeper understanding of pellet experiments in a reversed fieldpinch. In particular, evidence on the ablation cloud density, ablation rateand homogenization process are given by the measurements of twomid-infra-red interferometers located at different toroidal locations, one ofwhich is at the same poloidal plane as the pellet injector. For each pellet,the measurement of the latter interferometer displays a huge peak, which isdue to the crossing of the interferometer chord by the ablation cloud. Itsanalysis yields information on the cloud dimension and radial densitydistribution. The typical cloud density is much lower than that measured intokamaks. Due to such a low density, the stopping power of the ionized part ofthe ablation cloud is weak. As a result the ablation rate is higher than intokamaks. Another characteristic of the density increase measured by both of theinterferometers is the absence of dense plasma structures propagating for longdistances along field lines. This proves that the distance necessary for theablated material to become incorporated into the plasma is less than about 4 min the core of the discharge, due to a rapid mixing of the ablated material.Despite the different features displayed by pellet injection experiments inreversed-field pinches and tokamaks, the ablation and homogenization of thedeposited material can be described within the same general frame for bothmagnetic configurations. The differences arise mainly because of the differentmagnetic field topologies and values of the transport coefficients in the twomachines.

Study of Pellet-Plasma Interaction in the GAMMA 10 Tandem Mirror

In the GAMMA10 tandem mirror, the first pellet injection experiments in open systems have started.1 The tandem mirror machine is suitable for measuring propagation of pellet-fueled particles along the magnetic field lines. We observed a very low frequency oscillation in various diagnostic signals from the plasmas with a pellet injection. The frequency of the oscillation is a few hundred Hz. It is found that the phase of those signals changes to anti-phase at a certain place. The oscillation continued during plug-ECH for the formation of confining potential and its amplitude decays within a several ms after the end of ECH pulse. Measurements at several locations along the field lines indicates that this oscillation is a standing wave along field lines of plasma density perturbation due to the pellet injection. The wavelength of the oscillation is estimated to be about 6m. The oscillation has mode number n = 3.

Edge cooling experiments and non-local transport phenomena in ToreSupra

Edge cooling experiments have been performed in Tore Supra by impurityinjection and oblique pellet injection. Non-local transport (NLT) featureshave been confirmed in both cases in the ohmic regime. The pellet injectionexperiments have shown the existence of a threshold on the plasma density⟨ne⟩ and current Ip for the observation of thiseffect: ⟨ne⟩/Ip1/2⩽1.0×1016 (m-3 A-1/2). A second threshold on the relative density variationcaused by the pellet injection has also been observed: Δ⟨ne⟩/⟨ne⟩⩽0.5. It has been observed that inthe L regime during lower hybrid current drive (LHCD) experiments these limitsare extended: [⟨ne⟩/Ip1/2]crit = 1.4×1016 (m-3 A-1/2) and [Δ⟨ne⟩/⟨ne⟩]crit⩾0.65. Observations ina LHCD scheme with fully non-inductive plasma current seem to indicate thatthe mechanism governing the NLT is unlikely to be linked to the plasma currentprofile effects including the spatial redistribution, the magnetic shear, andthe low-m MHD modes.

Transient transport barriers in ohmic and electron cyclotron heated RTPplasmas: relation with rational magnetic surfaces

Oblique pellet injection experiments have been performed in RTP ohmic andelectron cyclotron (EC) heated plasmas. Peripheral cooling induces a transientrise of the electron temperature (Te) in the plasma core, with theformation of a region of a high Te gradient (transient transportbarrier). The position of the barrier appears to be linked with the location oflow-order rational magnetic surfaces. For Ohmic and EC heated plasmas withresonance moderately off-axis, the Te rise is peaked in the plasmacentre and the footpoint of the barrier is located close to the q = 2 surface.For EC plasmas with far off-axis resonance, i.e. in plasmas exhibiting hollowTe profiles, the Te rise starts off-axis, near the innermost loworder rational surface (typically q = 3). It then propagates to the centre in aseemingly diffusive way.

Damping of Surface Alfvén Wave in a Slab Plasma

The MHD wave is studied when two steep density gradient regions exist at surfaces of slab plasma. In such a case, it is shown that the surface Alfven wave has two branches with nearly the same damping rates, since the steep density gradients are located closely each other. However, for the sharp boundary plasma, the surface Alfven wave does not damp. As the density profile is relaxed, the damping rates become larger, pass via extremum, and again they become small when the scale length of the density gradients extremely increases. These damping rates seem consistent with behavior of magnetic fluctuations observed in the Heliotron-E pellet injection experiment.

Magnetohydrodynamic effects on pellet injection in tokamaks

The location at which pellets are injected into a plasma can have a significant effect on what fraction of the pellet mass remains in the plasma for refueling purposes. Magnetohydrodynamic (MHD) simulations presented here, confirm the results of pellet injection experiments: toroidal curvature makes it favorable to inject pellets from the inboard side or from the top or bottom, rather than from the outboard side. Sufficiently large pellets injected at the inboard edge can reach the plasma center, and in the process drive magnetic reconnection to produce negative central shear. Injection at the top (or bottom) of the tokamak causes relatively little displacement of the pellet. A scaling law is obtained for pellet displacement which agrees well with the simulations. The MHD simulations were carried out with a new unstructured mesh finite element version of the MH3D full MHD code.

Possible modification of Ohm's law due to radial plasma flux in the pellet injection experiments in the T-10 tokamak

In some experiments carried out in the T-10 tokamak, injection of macroparticles resulted in a particle flux directed inwards in an inner part of the plasma column. It is proposed to take into account the toroidal Lorentz electric field, induced by the radial transport plasma flow, which is derived from the experimental data on the particle balance, in Ohm's law. This approach allows an explanation of the loop voltage evolution in the pellet injection experiments in T-10

Thomson-scattering diagnostic on the Frascati tokamak upgrade

The Frascati tokamak upgrade Thomson-scattering system is used for the measurement of electron-temperature and electron-density spatial profiles along the vertical diameter of the tokamak at 19 spatial points up to 10 times in a single plasma discharge, with a spatial resolution that ranges from 2 cm in the central region to 4 cm in the plasma edge. The radiation source is a Nd:YLF laser that operates at 1053 nm, with a divergence of 0.4 mrad full angle, and is capable of delivering a burst of 10 pulses with energies of 4.5 J/pulse; the interpulse time can be regulated from 20 to 100 ms. The scattered radiation is collected by two objectives: the first looks at the plasma center, and the second at the plasma edge. Bundles of optical fibers in the focal plane of the objectives carry the scattered light from the tokamak hall to a set of 19 interference-filter polychromators, whose transmission is 70%, and the rejection of the stray light at the laser wavelength is 1/10(7). The detectors are avalanche photodiodes ith a noise-equivalent power of the order of 10(-13) W/(Hz)(½) at 1053 nm. The spectral calibration of the polychromators is presented. The absolute calibration of the scattering system for the electron-density measurement has been carried out by the use of Raman scattering on hydrogen and deuterium. Examples of the results of the temporal evolution of T(e) and n(e) spatial profiles are presented for ohmic plasma heating, lower-hybrid current drive, and a pellet-injection experiment. The electron-temperature and electron-density profiles measured through Thomson scattering are compared with the temperatures measured through the use of electron-cyclotron emission and the density profiles obtained from the interferometer data.