Research Case Study

Plasma Mg Evaporation Model

NH₃ plasma heating with coupled Electric Currents + Heat Transfer modeling across a 0–90 A current sweep.

COMSOL multiphysics study of magnesium disc heating and evaporation-risk behavior during plasma exposure. The model compares current loading and disc-thickness sensitivity to identify a more stable thermal operating window for nanoparticle synthesis.

My contribution: built the EC–HT sweep model, meshing strategy, evaporation-proxy boundary condition, and IR thermography validation workflow. Experimental reactor hardware belonged to my advisor’s lab; the modeling workflow is my work.

Electric potential and current distribution used to compute Joule heating — **Electrical solution:** current paths used to compute Joule heating and resulting thermal gradients.

Modeling workflow: EC–HT coupling · 0–90 A current sweep · evaporation-risk proxy · IR thermography comparison

Key result: Increasing Mg disc thickness reduced peak evaporation sensitivity at equivalent current, indicating a more stable plasma-heating regime.

Geometry + Mesh Evidence

The geometry and meshing strategy resolve high-current regions while keeping the full reactor domain computationally efficient.

Full assembly model — **Reactor domain assembly:** enclosure preserves realistic current return paths and boundary heat losses.

Finite element mesh — **Mesh refinement:** dense elements near contacts and high-current regions, coarser outer domain to control computational cost.

Model assumptions and limitations

Evaporation proxy: latent-heat sink boundary used as a relative Mg-loss indicator.
NH₃ plasma: represented through enclosure boundary conditions rather than full plasma chemistry.
Radiation: simplified Stefan–Boltzmann boundary treatment.
Contact resistance: treated as a sensitivity parameter affecting peak gradients.
Mesh convergence: checked to confirm key thermal trends were not dominated by element size.

Experimental Reactor Context

Experimental hardware provides physical context for the modeled geometry and operating constraints.

**Tungsten boat:** evaporation hardware used in plasma experiments.

Magnesium discs — **Mg discs:** thickness-varied samples used for evaporation tests.

Reactor T pipe — **Reactor T-junction:** gas and electrical feed location motivating the model domain.

Further lab system details and SOP contribution

The simulation work was grounded in a physical RF plasma reactor system used in the lab for Mg evaporation and nanoparticle synthesis. These images provide additional context behind the geometry, operating constraints, and hardware assumptions used in the model.

Context note: the experimental reactor hardware belonged to my advisor’s lab. I included these photos as supporting lab context for the model, while the simulation workflow, coupling strategy, meshing, and validation logic shown above are my work.

RF plasma system overview — **System overview:** RF plasma reactor with rotational drive, vacuum interface, and controlled gas path.

Electrode clearance and isolation check — **Interface checks:** electrode spacing and electrical isolation verified prior to plasma ignition.

Operational procedure: I authored a standard operating procedure for safe plasma system startup, RF ignition, gas stabilization, and shutdown to support repeatable nanoparticle synthesis experiments.

Reactor architectureQuartz-tube plasma reactor with rotational drive and furnace heating.
Electrical setupRF electrode coupling and grounding configuration.
Gas handlingHydrogen and argon flow control with vacuum stabilization.
Pre-ignition checksMechanical alignment and electrode-clearance verification.
Procedure controlStartup, gas stabilization, RF ignition, and shutdown documentation.
Validation supportIR thermography workflow used to compare model trends with observed heating behavior.

Selected References

Kortshagen et al., Nonthermal Plasma Synthesis of Nanocrystals, Chemical Reviews.
Wagner et al., Low-Temperature Plasma Hydrogenation of Mg Nanoparticles, Journal of Applied Physics.