Published papers

This paper numerically investigates the impact of neutral gas puffing on the divertor power exhaust and particle control in GAMMA 10/PDX applying the plasma fluid code LINDA and the kinetic neutral code KMNC. The LINDA code is applied to GAMMA 10/PDX plasmas for understanding the physics of detachment and energy loss processes. The LINDA code is coupled with a Monte–Carlo neutral code for describing the hydrogen (H and H2) neutral particles. The plasma ion temperature (Ti) is reduced towards the target plate with the increment of H2 Gas-Puff rate. This paper also numerically investigates the impact of recombination processes on the detached formation. The ionization, charge–exchange, and recombination of H have been included as a particle and an energy sink and source parts of the fluid equations. The reaction rate coefficient of molecular activated recombination (MAR) has been recently introduced in the LINDA code. In this paper, we study numerically the effect of MAR on the detached plasma. The MAR plays a dominant role in making the detached plasma. The simulation clarified that H neutral particles injection can effectively generate the detached plasma by enhancing the plasma-neutral collisions processes.

Result of hydrogen (H) and deuterium (D) experiments done by NIFS research and development negative ion source (RNIS) demonstrated that the co-extracted electron current with the negative ions and the electron density in the driver region in the D experiment have been around three times higher than that in the H experiment. To investigate mechanism of this difference, electron transport simulation using 3D kinetic particle tracking model KEIO-MARC code has been modified and applied to analysis of the isotope effect in the NIFS negative ion source. Simulation result suggests that impacts of isotope effects of sheath potential drop, coulomb collisions, and some reactions of ground state molecules and ions on the electron density is not large to explain the experimental result of the increase in the electron density in the plasma.

Sequences of hydrogen (H) and deuterium (D) experiments have been done by NIFS research and development negative ion source (RNIS) for the deuterium NBI development. In the experiments, the co-extracted electron current with the negative ions and the electron density in the plasma generation region in the D experiment have been around three times higher than that in the H experiment. To explain the difference of the electron density in the RNIS driver region, a zero-dimensional numerical model is developed in the present study. The model only focuses on the isotope effect for vibrationally excited level of electronically grounded state molecules and its relevant cross-sections. The calculation results show that difference of the ionization channel numbers via molecular vibrationally excited states could be a reason to enhance ionization rate in D plasma.

The emitting surface of an ion beam (plasma meniscus) in the vicinity of the extraction aperture is theoretically investigated. In the present study, the electrons, H+ ions and volume produced H− ions are contained in the electronegative plasma. By solving the one-dimensional Poisson equation, an analytical formula is obtained from the balance between the current reaching a wall and the space charge limited current of negatively charged particles. The analytical formula indicates that the distance, d eff, between the plasma meniscus and the extraction grid includes the ratio of masses of the negative ions and electrons, which is different from that for a plasma containing electrons and single positive ions, and depends on the ratio of the negative ion density to the electron density, α. With increase in the value of α, the distance d eff becomes shorter, and thus the penetration of the electric field for negative ion extraction into the source plasma becomes smaller. This tendency is confirmed by using a three-dimensional particle-in-cell simulation from the results of the contour map of the electrostatic potential near the plasma grid and the location of the plasma meniscus. The reasons why the distance d eff becomes shorter with increase in the value of α are considered to be the larger space charge effect of the negative ions compared with that of the electrons and the decrease in the plasma potential.

To extract intense ion beams with good beam optics from ion sources, controlling the distance deff between the plasma meniscus (i.e., beam emission surface) and the beam extraction grid is important. This study conducts a novel investigation into the dependence of the effective distance deff on the amount of surface H- production SH-. For this purpose, a 3D PIC (three dimensional Particle-in-Cell) simulation is conducted to obtain a model geometry of the extraction region for a H- ion source with SH- as a parameter. Based on results, deff significantly depends on SH- and the H--electron density ratio (α = nH-/ne) in front of the extraction aperture for the same plasma density; as SH- increases, deff decreases. The results suggest that SH- is critical for controlling deff and the resultant beam optics extracted from the negative ion source.

The power exhaust concept and an appropriate divertor design are common critical issues for tokamak DEMO design activities which have been carried out in Europe, Japan, China, Korea and the USA. Conventional divertor concepts and power exhaust studies for recent DEMO designs (Pfusion = 1 – 2 GW, Rp = 7 – 9 m) are reviewed from the viewpoints of the plasma physics issues and the divertor engineering design. Radiative cooling is a common approach for the power fusion scenario. Requirements on the main plasma radiation fraction (fradmain = Pradmain/Pheat) and the plasma performance constrain the divertor design concept. Different challenges contribute to optimizing the future DEMO designs: for example, (i) increasing the main plasma radiation fraction for ITER-level Psep/Rp designs and simplifying the divertor geometry, and (ii) extending ITER divertor geometry with increasing divertor radiation (Praddiv) for larger Psep/Rp ≥ 25MWm−1 designs. Power exhaust simulations with large Psep = 150 – 300 MW have been performed using integrated divertor codes considering an ITER-based divertor geometry with longer leg length (1.6 – 1.7 m), as in a common baseline design. Geometry effects (ITER like geometry or more open one without baffle) on the plasma detachment profile and the required divertor radiation fraction (fraddiv = Praddiv/Psep) were key aspects of these studies. All simulations showed that the divertor plasma detachment were extended widely across the target plate with a reduction in the peak heat load of qtarget ≤ 10 MWm−2 for the large fraddiv = 0.7 – 0.8, while the peak qtarget location and value were noticeably different in the partially detached divertor. Simulation results also demonstrated that radial diffusion coefficients of the heat and particle fluxes were critical parameters for DEMO divertor design, and that effects of plasma drifts on outboard-enhanced asymmetry of the heat flux, suggested the need for longer divertor leg to ensure the existence of a detached divertor operation with qtarget ≤ 10 MWm−2. Integrated design of the water cooled divertor target, cassette body (CB) and cooling pipe routing has been developed for each DEMO concept, based on the ITER-like tungsten monoblock (W-MB) with Cu-alloy cooling pipes. Engineering design adequate under higher neutron irradiation condition was required. Therefore, inlet coolant temperature (Tcool) was increased. In current designs, it still shows a large potential variation between 70 °C and 200 °C. The influence of thermal softening on the Cu-alloy (CuCrZr) pipe was fostered near the strike-point when the high qtarget of ∼10 MWm−2 was studied. Improved technologies for high heat flux components based on the ITER W-MB unit have been developed for EU-DEMO. Different coolant conditions (low- and high-Tcool) were provided for Cu-alloy and reduced activation ferritic martensitic (RAFM) steel heat sink units, respectively. The high-Tcool coolant was also considered for the CB and supporting structures. Appropriate conditions for the high-Tcool coolant, i.e. 180 °C/ 5 MPa (EU-DEMO) and 290 °C/ 15 MPa (JA-DEMO, CFETR and K-DEMO), will be determined in the future optimizations of the divertor and DEMO design.

In order to extract intense ion beams with good beam optics from hydrogen negative ion sources, it is important to control the shape of the plasma meniscus (i.e. beam emission surface). Recently, it is suggested experimentally that the degradation of beam optics in the RF negative ion sources may be due to the fluctuation of the distance d eff between the meniscus and the extraction grid caused by the fluctuation of the plasma density np . The purpose of this study is to make clear the dependence of d eff on np in the presence of a large amount of surface produced H- ions in order to understand such fluctuation of beam optics in RF sources For the purpose, 3D electrostatic PIC simulation was conducted taking the bulk plasma density as a parameter, investigating the extraction region of a H- ion source. A large amount of the surface H- production on the PG has been taken into account under the assumption that the H- production rate is proportional to the bulk plasma density. The result shows that the effective distance d eff is proportional to np-1/2 even for a large amount of surface H- production. This dependence suggests that the bulk plasma density np is the key parameters to control d eff and the resultant beam optics extracted from the negative ion source.

Temporal oscillation of the negative hydrogen ion (H-) beam in the Radio Frequency (RF) ion source is investigated by a simple 3D3V Particle-In-Cell (PIC) model. The model solves electron, proton, and H- transport processes in the extraction region. The calculation domain is from the vicinity of the beam aperture to the ground electrode. To understand the relation between the plasma density oscillation and the extracted H- beam characteristics, the input electron and proton fluxes from the driver region are varied parametrically with the fundamental J-PARC RF frequency (2 MHz). The numerical results give an idea of the main physical processes between the oscillations of the plasma parameters and the extracted H- ion trajectories depending on the plasma meniscus shape in the different RF phases.

Understanding plasma-material interaction is crucial for achieving steady-state operation of magnetic confinement fusion devices. Kinetic Monte Carlo (kMC) simulation is a powerful tool for investigating the motion of atoms in the plasma facing materials under the influence of this interaction. To predict trapping sites and migration energies necessary for kMC simulations, we developed a deep learning model based on pix2pix for predicting the spatial distribution of binding energy. Results show that the model can reproduce spatial distributions similar to the true values. However, larger errors occur in regions with steep value gradients.

Understanding the hydrogen recycling process is crucial for comprehending the behavior of detached plasma in nuclear fusion devices. To achieve this, a molecular dynamics (MD) model is being developed to predict the distribution of translational energies and rovibrational states of hydrogen atoms and molecules released from the plasma-facing materials. Neutral transport simulations, utilizing distributions obtained from the MD model as boundary conditions, are also a powerful tool for analyzing the impact of recycled hydrogens on edge plasma. However, the MD model requires significant computational resources to obtain distributions under varying material and irradiation conditions such as material temperature and incident energy. Therefore, developing effective models that seamlessly integrate neutral transport simulation with hydrogen recycling models is crucial. Machine learning techniques are employed to develop predictive models capable of forecasting distributions of energies and rovibrational states of released hydrogen atoms and molecules. Furthermore, a model considering the incident energy distribution (shifted-Maxwellian) is developed by integrating the monochromatic distribution with the shifted-Maxwellian distribution.

Construction of the JT-60SA tokamak was completed on schedule in March 2020. Manufacture and assembly of all the main tokamak components satisfied technical requirements, including dimensional accuracy and functional performances. Development of the plasma heating systems and diagnostics have also progressed, including the demonstration of the favourable electron cyclotron range of frequency (ECRF) transmission at multiple frequencies and the achievement of long sustainment of a high-energy intense negative ion beam. Development of all the tokamak operation control systems has been completed, together with an improved plasma equilibrium control scheme suitable for superconducting tokamaks including ITER. For preparation of the tokamak operation, plasma discharge scenarios have been established using this advanced equilibrium controller. Individual commissioning of the cryogenic system and the power supply system confirmed that these systems satisfy design requirements including operational schemes contributing directly to ITER, such as active control of heat load fluctuation of the cryoplant, which is essential for dynamic operation in superconducting tokamaks. The integrated commissioning (IC) is started by vacuum pumping of the vacuum vessel and cryostat, and then moved to cool-down of the tokamak and coil excitation tests. Transition to the super-conducting state was confirmed for all the TF, EF and CS coils. The TF coil current successfully reached 25.7 kA, which is the nominal operating current of the TF coil. For this nominal toroidal field of 2.25 T, ECRF was applied and an ECRF plasma was created. The IC was, however, suspended by an incident of over current of one of the superconducting equilibrium field coil and He leakage caused by insufficient voltage holding capability at a terminal joint of the coil. The unique importance of JT-60SA for H-mode and high-β steady-state plasma research has been confirmed using advanced integrated modellings. These experiences of assembly, IC and plasma operation of JT-60SA contribute to ITER risk mitigation and efficient implementation of ITER operation.

It is well known that the rate coefficient of molecular-assisted recombination (MAR) varies by several orders of magnitude depending on the rovibrational states of the hydrogen molecules. A molecular dynamics simulation is performed to estimate the rovibrational states of recycled hydrogen molecules at the divertor in the JA-DEMO reactor under detached plasma conditions. The simulation results reveal that molecules in high rovibrational states are released even with low incident energy, which will be the dominant condition under detached plasma conditions. Molecules generated in this way can strongly affect the formation of the detached plasma via a molecular assisted-process such as MAR.

Understanding of key physics and engineering parameters is one of the most important issues to conduct fundamental design studies of future fusion reactors. Also, sensitivity analyses of performance by parameter surveys need to be done for optimizing a reactor. This research aims at a qualitative analysis to identify directions of design parameter optimization for fusion reactors. For this purpose, a Core-SOL-Divertor (CSD) model has been employed, because it is suitable for a wide range of parameter surveys by changing various parameters with low computational costs. This paper analyzes the particle and energy balances of the JT-60U divertor plasma by using a CSD model and reveals dependencies of the balances on design parameters of the device. In addition, the same CSD model is applied to conduct divertor plasma analyses for a future demonstration reactor (JA-DEMO). The result shows that it is possible to obtain a low temperature state of its divertor plasma as has been already reported by a two-dimensional SOL-Divertor integrated code. From these results, this paper shows that the CSD model is applicable to basic studies of directions and design concepts for future fusion reactors.

For the first time, we have revealed the spatial structure of a detached plasma with electron–ion recombination (EIR) and molecular-activated recombination (MAR) in response to changes in ion temperature (Ti) and gas pressure. This was achieved using high-density sheet plasma generated by a linear divertor simulator (TPDsheet-ICR). The parallel-plate electrodes were positioned above and below the sheet plasma (∼1019 m−3). We varied the Ti from 3 to 7 eV using ion cyclotron resonance heating and investigated the parameters of the detached plasma near the target in the divergent field region. Plasma emission intensities at Balmer series (Hα and Hγ) and Fulcher band wavelengths (610 ± 10 nm) were measured with a high-speed camera equipped with an Arbaa prism. Additionally, a Langmuir probe measured the electron density and temperature of the plasma. Our results show that as the Ti increases, EIR generated near the plasma periphery disappears, while MAR gradually forms near the plasma center during the transition from detached to attached plasma. This transition occurs because the increase in Ti in the detached plasma transfers energy from ions to electrons, raising the high-energy component of the electrons, resulting in EIR disappearance and the onset of MAR.

We developed and performed a 3D Monte-Carlo simulation on local erosion and deposition in outer divertor target of JA-DEMO. The local erosion and deposition were evaluated considering the divertor plasma parameters and the roof shape of W-monoblocks. First, for local erosion simulation, the trajectories of D ions and impurity (He and Ar) ions were traced before irradiating the monoblock surface. The physical sputtering was calculated by these ions irradiation, considering their charge states, incident-energies and -angles. The highest sputtering flux was produced by multiply-charged Ar ions (Ar3+–Ar5+). The sputtering flux strongly depended on the poloidal positions along the divertor target. Although almost no erosion (sputtering) was obtained in low-temperature zone (∼1 eV), the erosion occurred in mid- (∼10 eV) and high-temperature (>10 eV) zone. Second, for local deposition simulation, the trajectories of W atoms emitted from the monoblock surface were traced. After transport (including ionization and recombination) in the plasma, W atoms and ions were deposited on the monoblock surface. The deposition occurred even on the area where plasmas did not irradiate (shadow). This was due to perpendicular diffusion to magnetic field lines, gyro-orbit effects for W ions and straight-line motion for neutral W atoms. The poloidal distribution of the deposition fluxes was similar to that of the sputtering fluxes. In low-temperature zone, no deposition flux occurred due to very low sputtering flux. The deposition flux in mid-temperature zone increased significantly and that in high-temperature zone remained high.

In a fusion DEMOnstration reactor (DEMO)-relevant scrape-off layer plasma (SOL), whose collisionality is lower than in the SOLs of present-day tokamaks, kinetic effects are predicted to reduce the plasma heat conductivity along the magnetic field below the value obtained with the classical Spitzer–Harm model. As a part of ongoing efforts to improve the predictive capability of SOL heat transport calculations, we have implemented the non-local heat flux model proposed by [Luciani, Mora and Virmont, Phys. Rev. Lett. 51 (1983) 1664–1667] (here referred to as the LMV model) in the integrated SOL–divertor simulation code SONIC in order to account for the kinetic effect on the heat conduction due to the parallel streaming of rapidly moving particles. Our transport simulations for the Japanese demonstration tokamak reactor concept, JA DEMO, show that the LMV model yields a significantly reduced ion parallel conductive heat flux density both on the low- and high-field side of the upstream SOL. The heat flux obtained with the LMV model seems to be consistent with earlier kinetic simulations of tokamak ion transport. As a consequence of the reduced heat flux, a significant increase in the ion temperature and a decrease in the density have also been found over a broad area of the upstream SOL.

In order to simulate hydrogen (H) plasma in the linear plasma device NAGDIS-II, we have modified the fluid code LINDA-NU to allow the simultaneous calculation of multiple ion species consisting of hydrogen atomic ions ((Formula presented.)) and molecular ions ((Formula presented.)). In this simulation, H and (Formula presented.) neutrals are assumed to be uniformly distributed in space in order to obtain initial qualitative results. The fraction of (Formula presented.) ions increases as the (Formula presented.) molecular density increases, and the recombination process between (Formula presented.) and electrons is observed to reduce the particle flux to the target plate. With an increase in H density, the electron density increases due to the decrease in ion flow velocity due to the change exchange process, and the electron temperature decreases to less than 1 eV, leading to the detached plasma formation attributed to the electron-ion recombination process.

Superconducting (SC) tokamak JT-60SA plays an essential role in fusion research and development by supporting and complementing the ITER project, providing directions to the DEMO design activity and fostering next generation scientists and engineers. Since the short circuit incident at the terminal joints of equilibrium field coil #1 during the integrated commissioning (IC) in March 2021, both EU and JA implementing agencies (IAs) have examined how to ensure safe operation of JT-60SA by mitigating the risk of possible discharge occurrence inside the cryostat. Based on the experience of the global Paschen tests, the IAs have established a strategy of risk mitigation measures, which is a combination of (i) reinforcement of insulation, (ii) avoiding unnecessary voltage application to the coil systems and (iii) immediate de-energization of the coils when deteriorated vacuum conditions are detected. Thanks to the considerable efforts of the Integrated Project Team members, the IC restarted in May 2023. After confirmation of the SC state of the coil systems (TF, EF and CS), the coil energization test and the plasma operation phase 1 (OP-1) started. The first plasma was successfully achieved on 23 October 2023 with a limited value of voltage and current applied to the coils. The plasma configuration control was also confirmed with low plasma current and low auxiliary heating power conditions. Based on the IO-F4E-QST collaboration, activities of JT-60SA have been shared with the IO and provided an important lesson for ITER assembly and commissioning, and will provide an outstanding contribution to fusion research at large. After OP-1, maintenance & enhancement phase 1 (M/E-1) starts from January 2024, in which in-vessel components are installed, and heating and diagnostic systems are extensively upgraded to allow a high power heating experiment planned in OP-2. In order to make the best use of JT-60SA, a newly organized JT-60SA experiment team will refine the research plan for the future high heating power operation phase.

One-dimensional SOL-divertor plasma fluid simulation code which considers anisotropy of ion temperature has been developed so as to deal with sheath theory self-consistently. In our fluid modeling, explicit use of boundary condition for Mach number M at divertor plate, e.g., M=1, becomes unnecessary. In order to deal with the Bohm condition and the sheath heat transmission factors at divertor plate self-consistently, we introduced a virtual divertor (VD) model which sets an artificial region beyond divertor plates and artificial sinks for particle, momentum and energy there to model the effects of the sheath region in front of the divertor plate. Validity of our fluid model with VD model is confirmed by showing that simulation results agree well with those from a kinetic code regarding the Bohm condition, ion temperature anisotropy and supersonic flow. We also show that the strength of artificial sinks in VD region does not affect profiles in plasma region at least in the steady state and that sheath heat transmission factors can be adjusted to theoretical values by VD model. Validity of viscous flux is also investigated.

In order to realize a feasible DEMO, we designed an in-vessel component including the conducting shell. The project is affiliated with the broader approach DEMO design activities and is conceptualized from a plasma vertical stability and engineering viewpoint. The dependence of the plasma vertical stability on the conducing shell parameters and the electromagnetic force at plasma disruption were investigated in numerical simulations (programmed in the 3D eddy current analysis code and a plasma position control code). The simulations assumed the actual shape and position of the vacuum vessel and in-vessel components. The plasma vertical stability was most effectively maintained by the double-loop shell.