The Particle-in-Cell Method
Particle-in-cell methods track sample particles moving through continuous fields defined on a grid, coupling kinetic motion to field solves.
Particles and fields together
The particle-in-cell (PIC) method models a plasma as a large collection of computational macro-particles, each representing many physical particles, moving in continuous phase space. Electromagnetic fields live on a fixed grid. The two are coupled in a cycle: deposit particle charge and current to the grid, solve the field equations on the grid, interpolate fields back to particle positions, and advance particle velocities and positions.
This hybrid avoids the O(N^2) cost of direct particle-particle forces by mediating all interactions through the grid, while still capturing kinetic effects that fluid models cannot, such as wave-particle resonance and non-Maxwellian distributions.
The four-step cycle
Each timestep performs charge/current deposition (scatter), a field solve (Poisson or Maxwell), field interpolation (gather), and a particle push, usually with the Boris algorithm for the magnetic rotation. The same interpolation weights are used for scatter and gather to avoid a self-force, a subtle but essential consistency condition.
Noise and stability
Because particles sample the distribution statistically, PIC carries discrete-particle noise that scales as one over the square root of particles per cell. The explicit method must resolve the plasma frequency and, for electromagnetic PIC, satisfy a CFL condition on the light-wave speed. Grid spacing must resolve the Debye length or a finite-grid instability heats the plasma.
- Scatter-solve-gather-push cycle per timestep
- Boris pusher for stable magnetized particle motion
- Statistical noise falls as 1/sqrt(particles per cell)
- Debye-length and plasma-frequency resolution constraints
PIC is central to first-principles kinetic modeling of plasma sources, sheaths, and instabilities relevant to both magnetic-confinement and direct-energy-conversion concepts.