Hydrogen Pellet Research Articles

Development of a fuseless small-bore railgun for injection of high-speed hydrogen pellets into magnetically confined plasmas

A fuseless, small-bore, two-stage railgun system particularly suited to accelerating frozen hydrogen pellets for fueling magnetically confined plasmas is described. This railgun system is a two-stage accelerator, with the first stage consisting of a combination of a hydrogen pellet generator and a gas gun, while the second stage is a railgun. It is a fuseless railgun in that the plasma armature is formed by electrically breaking down the propellant gas immediately behind the pellet. It has a bore size in the range of a few millimeters in diameter. Results are presented on hydrogen pellet acceleration, indicating that it is feasible to use a railgun to accelerate hydrogen pellets to high velocities; however, to achieve the highest hydrogen pellet velocity, one must reduce gun wall ablation (to minimize inertial and viscous drag) and deterioration of pellet integrity. Using a prototype system, hydrogen pellet velocities exceeding 2.8 km/s were achieved on a 2 m-long railgun for a cylindrical pellet of 3.2 mm diameter and 4 mm length.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Controls and diagnostics on a fuseless railgun for solid hydrogen pellet acceleration

A two-stage railgun system has been built, incorporating several controls and diagnostics, some including unique features to account for the fact that the projectile is a frozen hydrogen pellet for fusion reactor refueling. A timing circuit has been developed to monitor projectile breech and muzzle velocities and to automatically trigger a sequence of events critical for effective plasma armature railgun operation. This circuit can initiate electrical breakdown of the propellant gas directly behind an incoming projectile, thus enabling fuseless operation. It also triggers a streak camera and a flashlamp for photographing the arc and the outgoing projectile, respectively. The automatic timing circuit is expandable and has been extended to incorporate a trigger for transaugmentation. The timing circuit is immune to mistriggering due to electromagnetic interference or fragmentation of the fragile hydrogen pellets. >

Study of a new railgun configuration with perforated sidewalls

A novel railgun configuration with perforated sidewalls is investigated. The motivation for this configuration is the desire to minimize the detrimental effects of inertial and viscous drag at high velocities caused by the debris from the projectile and the gun wall trapped in the plasma armature. By creating perforations on the sidewalls, ablated material is continuously vented out from the railgun bore, minimizing inertial and viscous drag. The test has been done on a 1.2-m-long railgun with a 3.2-mm-diameter bore. Results for hydrogen pellet acceleration show that at high currents the perforated railgun outperforms the unperforated one. Combined with a newly designed cryogenic pellet generator and the first-stage gas gun, a solid hydrogen pellet velocity of 2.46 km/s has been achieved on the 1.2-m railgun.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Determination of the safety factor profile in TORE SUPRA from the largest striations observed during pellet ablation

The largest striations in Hα(Dα) light observed across the ablation cloud of injected hydrogen pellets can be explained in most cases by the presence of resonant magnetic surfaces. The authors suggest that it is possible to identify these striations and therefore to obtain a very detailed profile of the safety factor. This q profile exhibits shear plateaus related to rational q values, thus indicating the presence of magnetic turbulence

Optical visualization of electric and magnetic field perturbations in tokamak discharges by hydrogen pellet injection

Two-dimensional intensity distribution mappings of photographs of pellet ablation cloud trajectories in TFR and TORE SUPRA reveal irregular shapes of the luminous striations. The observed features are not well understood, but a likely interpretation is that these features are caused by pre-existing electric and/or magnetic field perturbations in the hot core of tokamak plasmas. It is suggested to further investigate pellet injection as a diagnostic tool for the study of plasma structures and transport phenomena

Pellet fueling system for ITER

Abstract The International Thermonuclear Experimental Reactor (ITER) will use an advanced, high-velocity pellet injection system to achieve and maintain ignited plasmas. Three pellet injectors are proposed. For ramp-up to ignition, a highly reliable, moderate-velocity (1- to 1.5-km/s) single-stage pneumatic injector and a high-velocity (1.5- to 5-km/s) two-stage pneumatic pellet injector that uses frozen hydrogenic pellets encased in sabots will be used. For the steady-state burn phase, a continuous, single-stage pneumatic injector is proposed; this injector will provide a flexible fueling source beyond the edge region to aid in decoupling the edge region (whose parameters are constrained by divertor requirements) from the high-temperature burning plasma. Because pellet fueling is more efficient than gas injection, it may be possible to use pellet fueling to provide all of the tritium and a portion of the deuterium needed to sustain the fusion reaction and to use deuterium gas injection to provide the required edge densities for optimum divertor conditions. In addition to decreasing the tritium throughput and raising the tritium burn fraction, this mode of operation allows a deuterium-rich boundary layer gas at the edge, which could significantly lower tritium inventories in plasma-facing components. All three pellet injectors are designed for operation with tritium feed gas. Issues such as performance, neutron activation of injector components, maintenance, design of the pellet injection vacuum line, gas loads to the reprocessing system, and equipment layout are discussed. Results and plans for supporting research and development in the areas of tritium pellet fabrication, continuous extruders, and high-velocity, repetitive two-stage pneumatic injectors are presented.

Laser-prearc railgun: Development for the application to a fuel pellet injector of a nuclear fusion reactor

The laser-prearc railgun, that utilizes the phenomenon of laser-induced arc formation, was constructed and tested with plastic pellet projectiles. We envision our railgun as especially well suited as a solid hydrogen pellet injector for magnetic confinement fusion. The system consisted of a gas gun for preacceleration of a pellet and a railgun for its primary acceleration. A Q-switched ruby laser was used to induce electrical breakdown of propellant helium gas behind a dielectric pellet in the railgun. The present railgun was shown to accelerate a plastic pellet up to a velocity of 2.4 km/s.

Central magnetohydrodynamic activity in pellet-fueled JT-60 plasmas

The central magnetohydrodynamic (MHD) activities are strongly affected by hydrogen pellet injection. The change in the sawtooth characteristics (viz., crash time, crash mechanisms, and sawtooth period) seem to be dependent on the density (and pressure) peakedness. With deepening pellet penetration, the sawtooth frequency becomes longer. At the sawtooth emerging after the deep pellet penetration into high-Ip limiter discharges, only a small amount of the central kinetic energy is released and the crash does not follow the fully reconnecting style. The sawtooth crash after the pellet injection tends to have more ideal-like characteristics for higher density and pressure peaking factors. At each sawtooth, the m=1 rotation frequency changes suddenly to the ion-diamagnetic direction or the codirection (parallel to the plasma current).

Authors

Authors

Development of a Two-Stage Pellet Injector for Heliotron-E

A two-stage pellet injector for Heliotron-E is constructed and tested. The aim is to increase pellet velocity for more flexible density profile control of the Heliotron-E plasma and also to conduct a pellet ablation study using a wider range of pellet velocity. The pellet velocity is limited to ∼1.4 km/s in the current six-pellet injector at Heliotron-E. The fundamental operation is simulated with the Quickgun code. The experimental results generally agree well (within 80 to 90%) with the code calculations. By using a newly developed high-pressure fast valve, a hydrogen pellet velocity of 3.2 km/s has been achieved without a supportive shell or sabot to protect the pellet, although more tests are needed to confirm whether pellets can reliably be accelerated to this high speed without fracturing. The dependence of the pellet velocity and breech pressure on the pump tube fill pressure is studied. The results show that the fill pressure is an important parameter. The effect of the clearance between the piston and the pump tube wall on the pellet velocity is also investigated. The wear and damage of the piston caused by the compressing propellant gas are investigated. It is shown that changes on the piston surface when hydrogen is used for fill gas are different from the case of helium.

Evolution of pellet clouds and cloud structures in magnetically confined plasmas

The subject of this study is the space and time evolution of initially low temperature high density particle clouds in magnetically confined hot plasmas, such as those produced by ablating cryogenic hydrogen pellets in fusion machines. Particular attention is given to such physical processes as heating of the cloud by the energy fluxes carried by incident plasma particles (classical flux limited energy transport by thermal electrons along the magnetic field lines and anomalous heat conduction across them), gas dynamic expansion with j⃗ × B⃗ produced deceleration in the transverse direction, finite rate ionization and recombination (collisional and radiative) processes, and magnetic field convection and diffusion. The results show the existence of a distinct structure in the ablatant cloud surrounding and ablating pellet: a hollow temperature profile coupled to a peaked density profile in the plane normal to the magnetic field direction. The separation distance between the high and low temperature and density layers is typically the ionization or confinement radius. Also the flutes developing preferentially at the cloud surface have, at a certain phase of their development, the same wavelength. The temperature and density variations from the cloud interior to the cloud periphery may exceed two orders of magnitude. The lifetime of this structure is measured on hydrodynamic time-scales.

Absolute calibration of a frozen hydrogen pellet mass detector

Fueling of plasma physics experiments by the injection of frozen hydrogenic pellets has been accomplished on many devices, including the Tokamak Fusion Test Reactor (TFTR), the Joint European Torus (JET), Tore Supra, and the Advance Toroidal Facility (ATF). Injection of pellets lead to density increases in the plasma with efficiencies varying from 50% to near 100%. Determination of the fueling efficiency requires knowledge of both the plasma density and the pellet mass. The pellet mass can be determined as the pellet passes through a microwave cavity by the change in power transmitted though the cavity. A one‐to‐one correspondence between the transmitted power and the pellet mass has been demonstrated, but this system must be calibrated by other techniques. The technique described here is measuring the pellet mass directly by measuring the pressure increase in a vacuum chamber into which the pellet is injected. A high‐sensitivity capacitance manometer is used to absolutely measure the pressure profile pulse during pellet injection. Since these fuel pellets are hydrogenic, absorption of hydrogen onto the chamber wall must be accounted for in the calibration procedure. A fast (1 ms response time) pressure manometer allows measurement of the pressure pulse profile from which the absorption can be estimated, thereby allowing a determination of the pellet mass and a calibration of the microwave cavity technique. Results are presented and compared with other pellet size diagnostics.

A Comparison of Hydrogenic Pellet Ablation Models with Experiment

This paper reports on several theories developed over the past 15 years to describe the ablation of a solid hydrogenic pellet injected into a hot plasma. The most widely accepted theory is the neutral gas shielding model. This model has been expanded to include ablation by fast ions (as well as electrons), realistic particle distribution functions, self-limiting ablation, and a cold ionized plasma shield beyond the ablating gas. Ablation measurements, including absolute pellet penetration and ablation profiles, from the Impurity Study Experiment, Poloidal Divertor Experiment, Doublet-III, Alcator-C, Tokamak Fontenay-aux-Roses, T-10, Texas Experimental Tokamak, Tokamak Fusion Test Reactor, and Joint European Torus experiments are compared with variations of the neutral gas shielding model under a range of input assumptions.

Ion thermal diffusion in the Texas experimental tokamak

Profiles of the ion thermal diffusion coefficient χi have been determined experimentally for several TEXT plasma discharges. In the confinement region, the inferred values of χi are larger than the values predicted by neoclassical theory by a factor of two to six. No evidence of a reduction of thermal losses through the ion channel is found after injection of hydrogen pellets. The inferred electron thermal conductivities are larger than the ion thermal conductivities by an average factor of three. The anomaly in χi is compared with transport coefficients predicted by different models that describe the ηi instability.

A microwave cavity for measurement of the mass of hydrogen pellets

A description is given of a nondestructive method utilizing a microwave cavity for measuring the mass of high-speed pellets of solid hydrogen. The cavity is designed for use on a multishot pellet injector, where eight pellets are fired successively with trajectories being parallel and symmetrical around the injector axis. The cavity is cylindrical with the axis coinciding with the injector axis. When a pellet passes through the cavity through holes of 15–16 mm diameter, the change in resonant frequency is proportional to the pellet mass. As a result of the cylindrical symmetry the sensitivity will be identical for all pellets. The frequency shift is measured directly and is converted to a signal proportional to the size of the pellet. The cavity was calibrated with pellets of H2 and D2 containing around 6×1020 atoms and with velocities between 1200 and 1500 m/s. The sensitivity was found to be 300±15 mV/1020 atoms in both cases. This is in fair agreement with estimates made from the dielectric constants of solid H2 and D2. The cavity is built together with two optical detectors for time of flight measurements to form an integrated diagnostic unit.

The ablatant state of a hydrogen pellet during its passage in a tokamak

By considering the ablatant to be a 4-species (H2, H, H+ and e−) reacting gas and assuming the validity of local thermodynamic equilibrium, the ablatant state of a hydrogen pellet during its passage in a tokamak is studied. Results of analysis showed that the pellet ablation rate is sensitive neither to the specific assumption regarding the variation of the ratio of the specific heats, γ in the ablated cloud, nor to the precise ablation process at the pellet surface. Except a proportional constant depending on the effect of dissociation and ionization, the scaling of the pellet ablation rate with respect to the plasma state and pellet radius is similar to that of the 1-species neutral shielding model; the overall ablation rate is reduced by approximately 15%.The ratio ri/rp of the ionization radius, ri to the pellet radius, rp is shown to be related to the pressure, pe and the equilibrium constant of ionization, Ki(Te) by pe/Ki(Te) of the plasma electrons, and approaches a constant value, (≈ 11), when pe/Ki(te) > 2 × 10−12. For combinations of pellet radius, plasma state and magnetic field strength of interest, irrespective of the ionization process, at an ablated cloud radius r ≈ 2rp, nearly 80% of the ambient electron energy flux is absorbed.Under some situations when there is a lack of balance between the energy input of the plasma electrons and the heating and expansion of the ablatant, a weak shock (flow Mach number, M = 1) can develop near the pellet during the expansion of the ablatant.

Pellet ablation profile measurements on the JT-60 tokamak

Four hydrogen pellets, two of 3 mm■×3 mml pellets and two of 4 mmF/×4 mml pellets, with the velocity of ∼2.3 km/s, were injected to OH and NB heated plasmas by a four-barrel pneumatic injector on the JT-60 tokamak. The pellet ablation profile was estimated by the measurement of the Hα emissions originating from the ablation cloud using a fiber optics array combined with the Hα -filtered multichannel photodiode. The fiber optics array consisting of ten channel bundles views the pellet pass in the vacuum vessel. The optical fiber was developed to tolerate baking up to 300 °C. The measured ablation profiles were compared with the calculated one by the neutral gas shielding model with a self-limiting effect. A good agreement was obtained between measured and calculated ablation profiles in OH plasmas; however, the fast-ion-induced ablation needed to be taken into account in the calculation for NB-heated plasmas.

Neoclassical analysis of impurity transport following transition to improved particle confinement

Strongly peaked impurity density profiles have been observed in Alcator C after injection of frozen hydrogen pellets. Recent experiments in TEXT, ASDEX, PBX, JET and TFTR have exhibited similar impurity accumulation during regimes of improved confinement. The paper presents calculations of the neoclassically predicted equilibrium profiles of intrinsic light and heavy impurities in Alcator C and light impurities in TEXT. These calculations were performed for comparison with experimentally determined peaked profiles observed after pellet fuelling. In both machines, carbon exists in the plateau collisionality regime and its transport is dominated by collisions with the hydrogen background ions and by temperature gradient effects. In Alcator C, molybdenum is in the Pfirsch-Schliiter regime and is driven mostly by collisions with carbon inside r/a ≃ 0.25 and by temperature gradients outside this radius. The full neoclassical multi-ion, mixed regime calculation required for the transport of carbon and molybdenum is presented. The predicted carbon profile in TEXT is in good agreement with observation; in Alcator C, less outward diffusion or an additional inward drift is required for the predicted carbon profile to agree with observation. The profile predicted for molybdenum (which may not be as close to equilibrium as carbon) is in fair agreement with observation. While the results of these studies do not support a rigorous claim of agreement with neoclassical impurity transport, the observed profiles for different regimes and experiments are consistently close to neoclassical-like peaking predictions.

Transport with pellet fuelling in the Texas Experimental Tokamak

In the Texas Experimental Tokamak, a discharge regime characterized by persistent peaked profiles was observed to be induced by pellet fuelling, and its transport properties were studied. The hallmark of the regime is the suppression of sawteeth, and the regime was attained by injecting hydrogen pellets to promptly double the plasma density. In each of several pairs of experiments, a pellet fuelled discharge was compared with an edge fuelled discharge with similar averaged electron density, plasma current and toroidal magnetic field in order to characterize the transport change and to look for causal changes in the plasma turbulence. The impurity, radiation and working gas particle profiles were more peaked for the pellet fuelled case. The values of rjt derived from measured ion temperature and density profiles for high density edge fuelled and pellet fuelled discharges indicate that ion pressure gradient driven turbulence should be reduced in the pellet fuelled case. The macroscopic effects were accompanied by microscopic changes. Measurements of turbulent density fluctuations in high density edge fuelled discharges give strong evidence that a component of the turbulence propagates in the ion diamagnetic direction and that this particular mode is reduced in pellet fuelled discharges. The effects of the reduction of an ion mode turbulence were sought in the energy confinement of the discharges, but it was found that for these experiments (tailored for the turbulence diagnostics) the energy flowing in the ion channel was not large enough to affect the energy confinement. Simulations were used to interpret some of the results. Discharge simulations which include the pellet injection can reproduce the sawtooth suppression. This result and known properties of discharges in which sawteeth are suppressed suggest that some of the profile effects (including peaking of the working gas particles) induced by pellet injection are due to sawtooth suppression. The particle peaking may cause the observed reduction in the turbulence which follows pellet injection.

Pneumatic pellet injector for JT-60.

The pneumatic 4-shot pellet injector has been installed and operated for JT-60 (JAERI Tokamak-60). The performance tests have proven that the device provides high speed pellets as planned. The maximum pellet velocity obtained in the hydrogen pellet tests is greater than 2.3 km/s at 100 bar propellant gas.