The model’s main test
A computer simulation recovered the characteristic frequencies of a trapped, electrically unbalanced plasma from noise in the modeled particles. The result appeared in longitudinal, transverse and full three-dimensional tests. An eigenfrequency is the natural oscillation frequency associated with a plasma mode. The study was designed to assess whether a spectral macroparticle simulation could reproduce those frequencies and describe the response of modeled eigenmodes to a rotating-wall drive.
The modeled system was a flat-top non-neutral plasma column in a cylindrical trap. It was represented either as an infinitely long periodic column or as a finite column bounded by mirrors. The numerical method advanced drift centers with drift-kinetic Vlasov dynamics and calculated the self-field with a Fourier–Bessel Poisson solver in a conducting cylinder.
Finding frequencies in noise
To make the frequency check, the researchers initialized macroparticle positions from a uniform spatial distribution and velocities from a centered Maxwellian distribution. They applied a fast Fourier transform, or FFT, to the complex mode-amplitude signals. Spectral peaks supplied the eigenfrequencies, while the full width at half maximum supplied the uncertainty estimate. The simulation’s initialization noise therefore served as the basis for the frequency spectrum.
The noise-based spectra recovered the reported dispersion relation in the longitudinal, transverse and full three-dimensional simulations. The comparison was close but not identical: expected frequencies were always slightly higher than the simulated values. The paper notes that diamagnetic drift was neglected in the comparison, but the analysis leaves unresolved whether that omission relates to the offset.
The comparison stayed within the modeled configurations. One used an infinitely long periodic column, while the other used a finite column with mirror boundaries. The evidence therefore supports the tested spectral scheme and these idealized configurations rather than every trapped-plasma condition.
What the rotating-wall runs showed
In rotating-wall runs, the simulated eigenmode amplitudes showed the expected drive–eigenmode beating, a small amount of additional jitter and damping of the initial transient. In the linear region, the time pattern agreed qualitatively with the analytical driven-mode expression. That expression assumed zero initial perturbation, while the simulations contained initialization noise.
The amplitude scan separated a proportional-response range from a nonlinear range within the model. At φRW /φ0 = 0.005, the driven response remained proportional at or below the reported threshold. Above it, another mode increased and the response became nonlinear. The study describes that pattern as higher-harmonic mode mixing. The threshold is a feature of the reported scan, not a general value for trapped plasmas.
Compression in the frequency scan
In the weak-drive frequency scan, eigenmode excitation coincided with enhanced compression near the plasma’s eigenfrequencies, while the response was smaller elsewhere. The mean kinetic-energy change reached up to 80% near an eigenfrequency. That figure belongs to the modeled weak-drive response; it does not establish long-term compression.
The longitudinal drive mode was another part of the comparison. With a single allowed longitudinal mode, matching modes were selectively excited. With multiple longitudinal modes, the simulations showed background compression, heating, shifted eigenfrequencies and a reduced response when the drive frequency was held fixed.
The result’s narrow operating range
The model omitted cooling mechanisms. External-drive runs exhibited heating, and the reported applicability was limited to short responses at low amplitudes. Long-term compression and strong-drive behavior were not simulated.
The study also used a flat-top plasma profile and idealized cylindrical geometry with periodic or mirror boundaries. Taken together, the evidence is numerical and model-based, supporting the tested scheme and configurations rather than every possible trapped-plasma condition.
The document is an arXiv version-one preprint dated 20 Aug 2026. The authors state that supporting data are available on reasonable request and that the simulation code is openly available.
The study’s conclusion is limited but specific: noise-based spectra recovered modeled eigenfrequencies, and weak-drive scans showed compression responses near those frequencies. The results do not establish long-term compression or strong-drive behavior, which were outside the simulations.
Paper data and sources
Original title: Simulation of trapped non-neutral plasma dynamics with rotating wall compression
Authors: Luisa Riik, Oliver Boine-Frankenheim
Journal/Repository: arXiv
Status: Preprint, not yet peer-reviewed
First online: 2026-08-20
DOI: Not available
Original paper · Full text