287

Papers

9/9

Roadblocks Active

3

Connections

A new pseudo orbit-averaging algorithm accelerates gyrokinetic plasma simulations by a factor of 30,000, making it feasible to compute the full kinetic equilibrium of a high-field mirror machine (the WHAM experiment at 17 Tesla) over collisional timescales. Previously, the nine-order-of-magnitude gap between fast particle transit times and slow collisional equilibration made such simulations prohibitively expensive. The computed equilibrium — including electrostatic potential and ion confinement time — matches analytic theory, validating the approach for HTS mirror fusion devices.

██████████ 0.9 hts-magnets Preprint

ReadSaveConnections

FIREFLY is a new software package that estimates how heat and particles hit divertor surfaces in fusion devices, using simplified transport models and Monte Carlo particle tracking coupled to the EIRENE neutral transport code. It is validated against high-fidelity EMC3-EIRENE simulations on the W7-X stellarator and runs far faster, enabling engineers to screen many divertor geometry options before committing to expensive detailed simulations. This directly addresses a bottleneck in designing components that must survive heat fluxes exceeding 10 MW/m² in reactor-scale devices.

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

ReadSaveConnections

This analytic theory study extends the physics of how external magnetic perturbations couple to resonant plasma layers, finding that ion parallel flow can shield the plasma from those perturbations in parameter regimes relevant to future large tokamaks. Resonant magnetic perturbations (RMPs) are a primary tool for suppressing edge-localized modes (ELMs) and avoiding disruptions, so knowing when plasma flows block their penetration determines how aggressively RMPs must be applied. The new predictions are validated against the SLAYER numerical code, providing a more complete picture than previous theories that neglected ion flow.

██████████ 0.9 plasma-disruption Preprint

ReadSaveConnections

Using bispectral analysis methods on experimental plasma data from a mirror device, this study demonstrates that energy flows nonlinearly from large-scale Rayleigh-Taylor instability modes into lower-frequency drift-wave modes — a cross-mode transfer that standard linear spectral analysis completely misses. The Ritz and Kim methods are validated on synthetic data first, then applied to real density fluctuation measurements at multiple radial positions to map the spatial structure of this coupling. Capturing such inter-mode energy pathways is a known gap in reduced turbulence transport models (like TGLF) used to design future fusion reactors.

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

ReadSaveConnections

A physics-informed neural network (PINN) learns to solve the drift kinetic equation for neoclassical toroidal viscosity (NTV) torque — a quantity that affects plasma rotation and stability — without requiring any labeled training data, by embedding the governing equations directly into its loss function. Validated against 20,000 solutions from the NTVTOK first-principles solver using EAST tokamak conditions, it achieves comparable accuracy at a fraction of the computational cost. Removing the need for a large simulation dataset is important because generating such data from first-principles codes is expensive and often impractical for real-time plasma control applications.

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

ReadSaveConnections

Gyrokinetic particle-in-cell simulations (using the ORB5 code on the standard ITPA benchmark case) show that when the linear drive rate exceeds about 0.47% of the wave frequency, Toroidal Alfvén Eigenmodes saturate due to thermal plasma nonlinearities rather than energetic particle effects — and the saturation amplitude is nearly independent of how hard the mode is driven. TAEs are important because they can expel the fast ions that heat the plasma in burning plasma experiments like ITER; understanding what limits their amplitude determines how much heating is lost. A complementary analytic theory explains the saturation via phase-space zonal structures.

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

ReadSaveConnections

Analytical theory and particle-in-cell simulations show that an externally injected right-hand circularly polarized (R-wave) can act as a 'firewall' that traps and decelerates runaway electrons at Doppler-shifted cyclotron resonance, preventing further acceleration. Runaway electrons are a serious disruption hazard in large tokamaks — they can damage the first wall if not controlled — and radio-frequency wave injection is a candidate mitigation technique; this work provides explicit physical mechanism and tokamak-relevant simulation evidence. The mechanism works through a pseudo-energy conservation that reverses the particle's parallel motion once it enters resonance with the wave.

██████████ 0.8 plasma-disruption Preprint

ReadSaveConnections

Laser-driven experiments at the LULI2000 facility, combined with 3D hybrid simulations, demonstrate that electron pressure anisotropy — not classical resistivity — is the dominant driver of tearing instability growth and plasmoid formation in magnetic reconnection. Magnetic reconnection underlies sawtooth crashes and major disruptions in tokamaks, events that can terminate the plasma prematurely; identifying the correct drive mechanism changes how reconnection rates and disruption onset thresholds should be modeled. The result also shows that resistivity and pressure isotropization act as stabilizers, pointing to new ways to think about disruption seeding.

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

ReadSaveConnections

This theoretical study shows that measurable geometric properties of the velocity and magnetic field gradient tensors — called tensor invariants — can serve as proxies for the directional energy fluxes in magnetohydrodynamic turbulence, with the purely hydrodynamic contribution expressible exactly in terms of those invariants. This matters for fusion because turbulent transport is what ultimately limits plasma confinement, and current reduced models cannot easily distinguish which physical mechanism (strain, vorticity, magnetic tension) dominates energy cascade at a given scale. The framework is illustrated with pseudospectral MHD simulations and provides a new diagnostic lens for both simulations and experimental data.

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

ReadSaveConnections

Standard magnetohydrodynamic (MHD) equilibrium theory, which works well for axisymmetric tokamaks, cannot produce smooth solutions in 3D toroidal geometries like stellarators because singular currents appear on resonant magnetic surfaces that violate the model's assumptions. The authors propose a statistical equilibrium framework based on ergodically fluctuating magnetic fields that avoids these singularities and yields smooth solutions, offering a physically self-consistent alternative for stellarator equilibrium calculations. This is relevant as stellarators like W7-X are increasingly viewed as disruption-free alternatives to tokamaks, but their equilibrium modeling has remained on less rigorous theoretical footing.

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

ReadSaveConnections