Verification of Electromagnetic Fully-kinetic Symplectic Particle-in-cell Method in Microinstabilities Simulation of Toroidal Plasmas

Jianyuan Xiao, Jian liu · 2025

Read Verification of Electromagnetic Fully-kinetic Symplectic Particle-in-cell Method in Microinstabilities Simulation of Toroidal Plasmas by Jianyuan Xiao, Jian liu (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-23112.txt.

Introduction

Jianyuan Xiao and Jian Liu test, in a 2025 physics paper, a full-kinetic code for small storms in ring-shaped fusion plasma. Their code tracks each charge kind in full, with field forces, and keeps a frame trait called symplectic form, which guards long-run truth. Set data came from the Cyclone Base Case, a known clash field. Three storm kinds were grown in the runs: ion heat slope storms, trapped electron storms, and balloon forms. Their growth rates matched those from lean gyro codes in past work.

The known calm of the first storm kind at high slope was caught as well. Shifts in the mass ratio and in the code's light speed moved rates just a touch. Full-kinetic cost may thus drop at the scales that count.

Catalog record

CREATORJianyuan Xiao, Jian liu
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE23 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-23112

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Verification of Electromagnetic Fully-kinetic Symplectic Particle-in-cell Method in Microinstabilities Simulation of Toroidal Plasmas Jianyuan Xiao, Jian liu Abstract We present a symplectic electromagnetic fully-kinetic particle-in-cell simulation of microinstabilities in plasma, using parameters from the Cyclone Base Case [Dimits, et al. , Physics of Plasmas 7, 969 (2000)]. The results show that the growth rates of unstable modes, including ion temperature gradient (ITG), trapped electron mode (TEM), and kinetic ballooning mode (KBM), are consistent with those obtained from previous gyrokinetic models. Additionally, the β \beta -stabilization of the ITG is reproduced. The analysis also reveals that the impact of the ion-electron mass ratio and the numerical speed of light on the growth rate of the most unstable modes is minimal. This suggests that the fully kinetic method offers a potential for reduced computational cost when investigating the physics of drift wave instabilities at relevant space-time scales. pacs 52.65.Rr, 52.35.Qz, 52.35.Ra, 52.25.Dg I Introduction It is widely believed that microinstability-driven turbulence is a key factor in the anomalous transport Horton 1999 of plasmas, which significantly impacts the performance of magnetic confinement fusion devices. These instabilities arise primarily from steep temperature and density gradients within the plasma. Typically, a gyrokinetic Lee 1983 ; Dubin et al. 1983 ; Brizard and Hahm 2007 approach is used to model this kind of turbulence. Previous electromagnetic gyrokinetic methods Pueschel et al. 2010 ; Holod and Lin 2013 ; Pueschel et al. 2008 have identified three types of microinstabilities at the ion motion scale, i.e., Ion Temperature Gradient (ITG), Trapped Electron Mode (TEM), and Kinetic Ballooning Modes (KBM) instabilities. The dominant unstable mode varies with plasma parameters, and simulations indicate that as plasma beta increases, these microinstabilities initially stabilize before becoming unstable once beta surpasses the so-called ballooning limit Chu et al. 1978 . Conventional gyrokinetic models simplify the evolution of both particle and electromagnetic fields compared to fully-kinetic charged particle-electromagnetic models, raising concerns about the impact of these simplifications on results. With the advent of 100 petaflop and exaflop supercomputers, it is now feasible to directly simulate microinstabilities in whole-volume magnetic fusion plasmas using fully-kinetic particle-in-cell (PIC) methods. This allows for verification of previous gyrokinetic models and the potential discovery of new physics that may not be captured by those models. However, traditional full-kinetic PIC methods face significant challenges when simulating charged particle-electromagnetic fields systems. First, the grid size must resolve the Debye radius ( \lambda_{\mathrm{D}} ), such as in the Boris-Yee scheme, to avoid self-heating from finite-sized grid instabilities Meyers et al. 2015 ; Barnes and Chacón 2021 ; Ueda et al. 1994 , which is problematic as the Debye radius in magnetic confinement fusion plasmas is typically much smaller than the global plasma size. For instance, with typical fusion plasma temperatures of T_{e}\sim 10\mathrm{keV} and density of 1\times 10^{20}\mathrm{m}^{-3} , the Debye radius is less than 0.1\mathrm{mm} , necessitating over 10,000 grids in each direction to resolve it in a typical plasma size about 1\mathrm{m} , leading to computational demands that are currently unmanageable. Additionally, the time-step in conventional explicit PIC schemes is limited by the Courant-Friedrichs-Lewy (CFL) condition and plasma frequency ( \omega_{\mathrm{pe}} ), requiring over a million time-steps to resolve one period of unstable modes. This long-term simulation can accumulate significant numerical errors, compromising result reliability. To address these issues, we employed a newly developed explicit 2nd-order charge-conservative symplectic structure-preserving electromagnetic fully-kinetic (FK) PIC scheme for cylindrical geometry Jianyuan and Hong 2021 ; Xiao et al. 2021 to simulate microinstabilities using the established Cyclone Base Case (CBC) Dimits et al. 2000 parameters. This method eliminates the self-heating problem, and its superior long-term conservative properties ensure result accuracy. We identify frequency and growth rate of unstable modes and our findings include: 1) The growth rates of ITG and TEM from the FK method are slightly higher than predictions from gyrokinetic models, while the KBM growth rate is comparable. 2) The mode frequencies for ITG and TEM align closely with gyrokinetic predictions, whereas the KBM frequency is lower than expected. 3) The growth rate of the most unstable mode initially decreases with increasing plasma beta for ITG and TEM, while it significantly increases for KBM, consistent with previous electromagnetic gyrokinetic simulations. This paper is organized as follows. Section 2 provides a brief introduction to the electromagnetic fully kinetic physical model and the symplectic structure-preserving method. Section 3 details the construction of Tokamak equilibrium, including Cyclone Base Case parameters and simulation results. Finally, Section 4 concludes the paper. II Symplectic Electromagnetic fully-kinetic particle-in-cell scheme employed in SymPIC First, we briefly introduce the symplectic electromagnetic fully-kinetic PIC scheme in the cylindrical coordinate developed previously Jianyuan and Hong 2021 which is implemented in SymPIC code and it can run efficiently on modern heterogeneous supercomputers with up to over 40 million cores Xiao et al. 2021 . Previously, we have applied it to simulate Tokamak edge plasmas and obtained comparable results with fluid simulations and experiments Liu et al. 2022 . In this work, we focus on verifying of simulations of microinstabilties in core plasmas using this approach. This work focuses on the physical model of charged particles coupled with electromagnetic fields, of which the evolution equations are \displaystyle\ddot{\mathbf{x}}_{s,p} \displaystyle= \displaystyle\frac{q_{s}}{m_{s}}\left(\mathbf{E}\left(\mathbf{x}_{s,p}\right)+\dot{\mathbf{x}}_{s,p}\times\mathbf{B}\left(\mathbf{x}_{s,p}\right)\right)\penalty\ , (1) \displaystyle\dot{\mathbf{E}} \displaystyle= \displaystyle\nabla\times\mathbf{B}-\sum_{s,p}q_{s}\dot{\mathbf{x}}_{s,p}\delta\left(\mathbf{x}-\mathbf{x}_{s,p}\right)\penalty\ , (2) \displaystyle\dot{\mathbf{B}} \displaystyle= \displaystyle-\nabla\times\mathbf{E}\penalty\ , (3) where \mathbf{x}_{s,p},m_{s},q_{s} indicates the location, mass, charge of the p -th particle of the s -species, \mathbf{E} and \mathbf{B} are electromagnetic fields. For simplicity, we normalize both permittivity \epsilon_{0} and permeability \mu_{0} in the vacuum to 1. To build symplectic structure preserving PIC method in the cylindical geometry, we start from the action integral of the charged particle coupled with electromagnetic fields system, which reads \displaystyle\mathcal{A}[\mathbf{x}_{sp},\dot{\mathbf{x}}_{sp},\mathbf{A},\phi] \displaystyle= \displaystyle\int\mathrm{d}t\sum_{s,p}\left(L_{sp}\left(m_{s},\mathbf{v}_{sp}\right)+q_{s}\left(\mathbf{v}_{sp}\cdot\mathbf{A}\left(\mathbf{x}_{sp},t\right)-\phi\left(\mathbf{x}_{sp},t\right)\right)\right) (4) \displaystyle+\int\mathrm{d}V\mathrm{d}t\frac{1}{2}\left(\left(-\dot{\mathbf{A}}\left(\mathbf{x},t\right)-\nabla\phi\left(\mathbf{x},t\right)\right)^{2}-\left(\nabla\times\mathbf{A}\left(\mathbf{x},t\right)\right)^{2}\right)\penalty\ , where \mathbf{x}=[x_{1},x_{2},x_{3}] and \mathrm{d}V=|h_{1}\left(\mathbf{x}\right)h_{2}\left(\mathbf{x}\right)h_{3}\left(\mathbf{x}\right)|\mathrm{d}x_{1}\mathrm{d}x_{2}\mathrm{d}x_{3} . In cylindrical coordinate, we choose line elements as \displaystyle h_{1}\left(x_{1},x_{2},x_{3}\right) \displaystyle= \displaystyle 1\penalty\ , (5) \displaystyle h_{2}\left(x_{1},x_{2},x_{3}\right) \displaystyle= \displaystyle(x_{1}/X_{0}+1)\Delta x_{2}/\Delta x_{1}\penalty\ , (6) \displaystyle h_{3}\left(x_{1},x_{2},x_{3}\right) \displaystyle= \displaystyle\Delta x_{3}/\Delta x_{1}\penalty\ . (7) Here \Delta x_{1},\Delta x_{2} and \Delta x_{3} are reference grid size for each direction, X_{0} is the location of inner boundary of the simulation domain. Hence the velocity of each particle is \displaystyle\mathbf{v}_{sp} \displaystyle= \displaystyle\left[\dot{x}_{1,sp}h_{1},\dot{x}_{2,sp}h_{2},\dot{x}_{3,sp}h_{3}\right]\penalty\ , (8) where \displaystyle\left[x_{1,sp},x_{2,sp},x_{3,sp}\right]=\mathbf{x}_{sp} (9) is the coordinate of each particle in the cylindrical geometry. Finally the Lagrangian for each particle is \displaystyle L_{sp}\left(m_{s},\mathbf{v}_{sp}\right) \displaystyle= \displaystyle\frac{1}{2}m_{s}\mathbf{v}_{sp}^{2}, (10) and the electromagnetic fields are defined as \displaystyle\mathbf{E} \displaystyle= \displaystyle-\dot{\mathbf{A}}-\nabla\phi\penalty\ , (11) \displaystyle\mathbf{B} \displaystyle= \displaystyle\nabla\times\mathbf{A}\penalty\ . (12) The evolution equations Eqs. ( 1 )-( 3 ) can be derived from variation, i.e., \displaystyle\frac{\delta\mathcal{A}}{\delta\mathbf{x}_{sp}} \displaystyle= \displaystyle 0\penalty\ , (13) \displaystyle\frac{\delta\mathcal{A}}{\delta\mathbf{A}} \displaystyle= \displaystyle 0\penalty\ , (14) \displaystyle\frac{\delta\mathcal{A}}{\delta\phi} \displaystyle= \displaystyle 0\penalty\ , (15) and the key to build symplectic algorithm is to find the discrete action integral. According to our previous work Xiao et al. 2021 , the action integral can be discretized using Discrete Exterior Calculus (DEC), Whitney interpolating form and zigzag integral path of particles to guarantee the discrete gauge invariance property. In such discretization, local charge conservation property is automatically satisfied and hence we can directly use electromagnetic fields rather than potentials to build the algorithm. Then the final explicit iteration scheme can be obtained by using discrete variational method or Hamiltonian splitting technique. Detail of the 2nd-order charge-conservative symplectic PIC scheme for the cylindrical coordinate

Read the complete text