285

Papers

9/9

Roadblocks Active

1

Connections

Using a novel pseudo orbit-averaging algorithm, researchers achieved a 30,000x speedup in gyrokinetic simulations of the WHAM HTS mirror experiment (17T, mirror ratio 32), allowing full kinetic equilibrium calculations over timescales previously inaccessible. The simulated ion confinement times and electrostatic potentials matched analytic predictions, validating the approach. This matters because gyrokinetic codes are normally too computationally expensive for routine fusion design work; this breakthrough could make first-principles turbulence and transport modeling practical in the design loop.

██████████ 0.9 turbulence-modeling Preprint

ReadSaveConnections

FIREFLY is a new computational package that couples simplified heat transport modeling to the EIRENE neutral particle code, enabling fast evaluation of divertor geometries without running expensive full plasma simulations. It was demonstrated on the W7-X stellarator and supports parameter scans and multivariate optimization of divertor shapes. Divertor design is a critical bottleneck for future reactors — the exhaust of heat and particles must be managed to prevent plasma-facing component damage, and faster design iteration tools directly accelerate progress on this problem.

██████████ 0.9 divertor-thermal Preprint

ReadSaveConnections

Gyrokinetic particle-in-cell simulations using the ORB5 code show that toroidal Alfvén eigenmodes (TAEs) — waves driven by energetic fusion-born alpha particles — saturate at amplitudes governed by thermal plasma nonlinearities rather than by the drive strength alone, above a threshold of gamma_L/omega_n > 0.47%. This 'stiffness' means TAE amplitude is relatively insensitive to how hard the mode is driven, which has implications for predicting alpha particle losses in burning plasmas like ITER. The frequency decrease at higher amplitude, caused by phase-space zonal structures in the thermal plasma, provides a measurable signature that diagnostics could track.

██████████ 0.8 long-confinement Preprint

ReadSaveConnections

This paper extends analytic boundary layer theory to include ion parallel flow at plasma resonant surfaces, deriving a modified stability index that captures how ions shield the plasma against externally applied magnetic perturbations used for ELM control. Numerical validation with the SLAYER/GPEC code confirms the analytic predictions. This matters because applied magnetic perturbations are the leading method for suppressing ELMs (edge-localized modes) in ITER, and accurately modeling the shielding response determines whether these perturbations will work at ITER-relevant plasma parameters.

██████████ 0.8 elm-control Preprint

ReadSaveConnections

Using the Ritz and Kim bispectral methods, this study maps energy transfer among Rayleigh-Taylor and drift-wave modes in plasma density fluctuations, finding that energy flows from faster RT modes into lower-frequency drift-wave modes through quadratic (three-wave) coupling. The analysis also identifies when these methods are statistically valid — specifically, they require data with near-Gaussian statistics and spatial stationarity. Understanding how turbulent energy cascades between mode types is essential for building predictive transport models that can forecast confinement quality in future reactors.

██████████ 0.8 turbulence-modeling Preprint

ReadSaveConnections

A physics-informed neural network trained without any labeled solution data can solve the drift kinetic equation and accurately reproduce neoclassical toroidal viscosity (NTV) torque profiles in tokamaks, validated against 20,000 EAST experiment cases. The key advance is using the governing physics equations as the training signal rather than pre-computed solutions, which avoids the cost of generating large labeled datasets. NTV torque controls plasma rotation, which in turn affects stability and confinement — faster NTV calculations could enable real-time plasma rotation control in future devices.

██████████ 0.8 turbulence-modeling Preprint

ReadSaveConnections

Laser-driven experiments at the LULI2000 facility, supported by 3D hybrid simulations, show that anisotropic electron pressure — not classical resistivity — is the primary driver of tearing instability growth rates in plasmoid-mediated magnetic reconnection. The experiments created magnetized counterflowing plasmas with anti-parallel field geometries similar to current sheet configurations in fusion disruptions. Magnetic reconnection is a key mechanism underlying disruptions and sawtooth crashes in tokamaks; understanding what controls its rate and onset is important for predicting and managing these events.

██████████ 0.7 plasma-disruption Preprint

ReadSaveConnections

This paper establishes that scalar invariants of coarse-grained velocity and magnetic field gradient tensors can serve as reliable proxies for energy flux across scales in MHD turbulence, with exact bounds derived analytically and validated in pseudospectral simulations. The approach provides a computationally cheap way to estimate turbulent energy transfer without resolving every scale — relevant because turbulent energy transport governs heat losses in fusion plasmas. The analytic bounds clarify precisely when and why these proxies break down, giving practitioners a way to assess their validity in specific plasma conditions.

██████████ 0.7 turbulence-modeling Preprint

ReadSaveConnections

Particle-in-cell simulations on a 1600-cubed grid show that three-dimensional effects systematically delay the onset of magnetic reconnection compared to 2D models, and that a strong guide field further delays onset by reducing linear growth rates and causing oblique modes to decohere. Despite this delay, all runs ultimately enter a fast-reconnection phase at a normalized rate of 0.08–0.10, independent of the 2D/3D geometry. This matters for disruption physics: if 3D effects delay reconnection onset in real tokamaks, disruption timelines may be longer than 2D simulations predict, potentially giving more time for mitigation systems to act.

██████████ 0.7 plasma-disruption Preprint

ReadSaveConnections

Combining analytical trajectory analysis with particle-in-cell simulations, this study finds that particles accelerated by a parallel electric field become trapped at Doppler-shifted cyclotron resonance with a circularly polarized wave, after which they experience deceleration along their original direction simultaneously with strong perpendicular acceleration — a counterintuitive 'firewall' effect. This mechanism is relevant to wave-based heating schemes in fusion devices (e.g., ICRH), where understanding how electromagnetic waves interact with accelerating particles in complex field geometries affects heating efficiency and can lead to parasitic particle losses.

██████████ 0.7 plasma-disruption Preprint

ReadSaveConnections