Since 18 of December 2019 uses Nucleus credentials. Visit our help pages for information on how to Register and Sign-in using Nucleus.
May 10 – 15, 2021
Virtual Event
Europe/Vienna timezone
The Conference will be held virtually from 10-15 May 2021

Simulation of plasma and neutral particles during H gas puffing in the divertor region of GAMMA 10/PDX using the fluid and kinetic neutral code

May 12, 2021, 2:00 PM
4h 45m
Virtual Event

Virtual Event

Regular Poster Magnetic Fusion Theory and Modelling P4 Posters 4


Dr Md Shahinul Islam (National Institute for Fusion Science, Japan)


We have recently developed a kinetic neutral code for understanding the hydrogen neutral gas dynamics in the divertor region. The Linear Divertor Analysis with fluid model (LINDA) code [A] has been successfully coupled with Kinetic Monte-Carlo Neutral Code (KMNC) in the present study. This is the first attempt for the GAMMA 10/PDX detached plasmas to perform such a self-consistent and detailed analysis using the fluid plasma and kinetic neutral code. We then investigate numerically the plasma and neutral particles transport behavior at the end-cell of GAMMA 10/PDX. The simulation results clarified that hydrogen (H) neutral particles can effectively reduce the plasma energy and can generate the detached plasma by enhancing the ionization, the charge-exchange, and the recombination loss. The obtained knowledge from the study contributes to understanding the physical mechanism of plasma detachment operation by H neutral particles in the divertor region of the fusion devices.

Fusion reactors cannot be realized without reducing and/or controlling the heat and particle loads to the divertor plate by plasma detachment. Therefore, understanding of the plasma detachment mechanism and of roles/effects of neutral particles on the detachment processes are indispensable in order to realize the long pulse operation of fusion reactors. In particular, the impact of atomic and molecular processes on the detachment is an unsolved critical research issue. Linear plasma devices greatly contribute to revealing plasma physics because the linear device has a much more simple structure than tokamaks. The impact of H Gas-Puffing (GP) on the plasma parameters has been investigated in the divertor region of the linear plasma confinement device GAMMA 10/PDX [B]. However, the physical mechanism of plasma detachment and energy loss processes during plasma-neutral interactions have not yet been clarified so far. In particular, the impact of atomic and molecular processes of neutral particles on the energy loss processes has not yet been understood. Consequently, the numerical simulation research has been started in order to clarify unsolved physical issues noted above.

In this simulation, the LINDA code and the KMNC code have been numerically solved in the self-consistent manner. More specifically, plasma transport is defined by the fluid equations while the neutral particles are defined by the kinetic neutral code. The recycling and puffed neutral particles are considered as $H_2$. The test particles are puffed into the bulk plasma with the cosine distribution. By applying a newly developed integrated code, LINDA+KMNC, we found that neutral particles concentrate near the target plate and reduce towards the upstream region. It was also found that the ion temperature ($T_i$) is reduced in the vicinity of the target plate when H$_2$ Gas-Puff (GP) is performed. The plasma ion density ($n_i$) is also increased near the target plate by the GP. The charge-exchange (CX) loss and the momentum loss increase significantly near to the target plate when GP is performed, which induces a significant reduction in the $T_i$. Moreover, the $n_i$ is increased while electron temperature ($T_e$) is reduced by the ionization of neutral particles. At the higher neutral injection cases, $T_e$ is reduced to below the threshold of ionization potential. Consequently, the $n_i$ is reduced by the recombination. These simulation results may solve the critical issues such as the role of atomic processes on the energy loss processes and on the detached plasma formation.

Mesh structure of the model.
2D profiles of H atom neutral density calculated by the kinetic neutral code (a) Case a: without GP, (b) Case b: moderate GP, and (c) Case c: high GP.

In the LINDA code, 2-D numerical grids have been created in the cylindrical coordinate system based on the axisymmetric condition. Figure 1 shows the mesh structure of the simulation region. A tungsten (W) target plate is designed at the end of the simulation space, as shown in Fig.1. Figure 2 shows the H neutral atom density ($n_{oH}$). For the case a (without GP), the neutral density comes from the recycling neutral particles and is shown to be very small. However, the $n_{oH}$ is enhanced significantly near the target plate when H$_2$ GP is performed as shown in Fig. 2(b)-(c). The $n_{oH}$ increases with increasing GP rate.

The axial profiles of $T_i$, $n_i$, and source terms of the fluid equations are shown in Fig. 3(a)-(d). The cases of the Fig.3 legends are explained in the Fig.2 caption. For without GP, the $T_i$ profile reduces slightly. Furthermore, the $T_i$ reduces drastically in the direction of the target plate when H$_2$ GP is performed. As shown in Fig. 3(b), the $n_i$ increases significantly near the target plate. Since neutral density is higher closer to the target plate, as shown in Fig. 2(b-c), the energy loss terms are also concentrated close to the target plate. As shown in Fig. 3(c)-(d), the ion energy and momentum loss are greatly enhanced, which leads to a strong reduction in $T_i$.

Axial profile ($r$ = 0) of (a) $T_i$, (b) $n_i$, source and sink terms of the (c) ion energy, and (d) momentum equation.

The fluid code has been compared with the Particle-In-Cell (PIC) code [C] for the first time to understand the validity of the fluid code LINDA. As shown in Fig. 4, the $x$-axial profile of $n_i$ and $u_{i||}$ of the two codes are found to be consistent, which indicates the validity of the LINDA code.

Axial profile of $n_i$ and $u_{i||}$ (a) LINDA code, and (b) PIC code.
Finally, the LINDA code outcomes are compared with the experimental data and plotted in Fig. 5. In particular, the pattern of the allover in the $n_e$ qualitatively agrees with the simulation outcomes. In addition, for higher neutral injection cases, the $T_e$ is saturated for both the simulation and the experiment.
Dependence of $n_e$ and $T_e$ (a) simulation, and (b) experiment as a function of H gas throughput.

[A] M.S. Islam et al., Plasma Phys. Control. Fusion 59 (2017) 125010.
[B] M.S. Islam et al., Plasma Fusion Res. 11, 2402042 (2016).
[C] T. Pianpanit et al., Plasma Fusion Res. 11, 2403040 (2016).

Affiliation National Institute for Fusion Science (NIFS)
Country or International Organization Japan

Primary authors

Dr Md Shahinul Islam (National Institute for Fusion Science, Japan) Prof. Yousuke Nakashima (Plasma Research Center, University of Tsukuba) Prof. Seiji Ishiguro (National Institute for Fusion Science ) Prof. Hatayama Akiyoshi (Keio University) Prof. Kazuo Hoshino (Keio University) Dr Hiroki Hasegawa (National Institute for Fusion Science) Prof. Naomichi Ezumi (University of Tsukuba) Prof. Mizuki Sakamoto (Plasma Research Center, JpUTsukuba)

Presentation materials