Research

Our research spans three connected directions: the fundamental physics of magnetized plasmas, applications to fusion and space technology, and interdisciplinary work joining kinetic plasma theory with gravitation and scientific machine learning.

01 · Understanding plasma

Fundamental plasma physics

We investigate how magnetized plasmas generate and reorganize fields, convert energy, and approach—or remain far from—equilibrium. Our canonical-vorticity framework treats magnetic field and fluid vorticity as parts of a single dynamical quantity, providing a unified view of magnetic reconnection, magnetogenesis, and related vorticity dynamics.

At kinetic scales, we connect single-particle trajectories to evolving phase-space distributions and then to macroscopic moments such as density, flow, current, and temperature. This micro-to-macro approach underpins our work on non-equilibrium relaxation, plasma thermodynamics, and coherent wave–particle interactions.

  • Magnetic reconnection
  • Magnetogenesis
  • Canonical-vorticity dynamics (QHD)
  • Non-equilibrium dynamics
  • Plasma thermodynamics
  • Wave–particle interactions

Selected papers

A disequilibrated Harris current sheet with a weak guide field relaxes into a mixed equilibrium with a locally amplified guide field
A disequilibrated Harris current sheet relaxes into a mixed equilibrium while locally amplifying its guide field. Cropped from Yoon et al., 2023, Fig. 1d. CC BY 4.0.
02 · Putting plasma to work

Applied plasma physics

We translate fundamental plasma processes into strategies for fusion energy, spacecraft propulsion, and space-environment forecasting. In magnetic-confinement fusion, we study wave activity, resonant scattering, and the firewall effect as routes to controlling energetic and runaway electrons. We also investigate reconnection and beam-driven turbulence in compact, non-inductively formed plasma configurations relevant to novel fusion concepts.

For space applications, we use reconnection and coherent wave–particle dynamics to improve the physical basis for space-weather prediction and control. We also explore electric-propulsion concepts in which asymmetric magnetic reconnection converts normally bidirectional Alfvénic exhaust into net thrust.

  • Nuclear fusion
  • Runaway suppression
  • Novel fusion reactor concepts
  • Space propulsion
  • Space-weather prediction and control

Selected papers and projects

Three-dimensional particle-in-cell simulations compare stable and merging beam-driven flux ropes using current-density volumes, magnetic field lines, and electron drift velocity maps
Particle-in-cell simulations comparing stable and merging beam-driven flux ropes through current density, field topology, and electron drift velocity. Park et al., 2025, Fig. 4. CC BY-NC-ND 4.0.
03 · Across disciplines

Interdisciplinary

We combine plasma kinetic theory with ideas from gravitation and data science. The Einstein–Vlasov framework links collisionless phase-space dynamics to evolving spacetime geometry, while gravitoelectromagnetism provides a carefully bounded analogy between electromagnetic fields and gravitational inertial or tidal effects.

Our scientific-machine-learning efforts use physics-informed neural networks (PINNs) to reconstruct canonical-vorticity fields and non-ideal terms from sparse spacecraft observations. A complementary direction develops data-driven closures that carry kinetic information into reduced plasma-fluid simulations without the full cost of a particle model.

  • Gravitoelectromagnetism and Einstein–Vlasov theory
  • PINN-integrated data reconstruction
  • Plasma simulation with data-driven closures

Frameworks and current directions

Green translucent phase-space surfaces and two dense cores in a gravity-focused numerical visualization
In-house visualization for ongoing gravity-focused numerical work.