All projects
MAE 6540 Plasmas for Propulsion, Cornell University

Microwave Plasma Jet Engine Simulation

A 3D finite-difference model of atmospheric microwave plasma ignition

Role
Sole author
Status
Completed
Microwave Plasma Jet Engine Simulation

Headline results

0.25 µs
Simulated time to peak electron temperature
N₂ + O₂
Species with separate impact-ionization thresholds
3D
Finite-difference spatial field, marched in time
MATLABFinite DifferencePlasma PhysicsCollision Cross SectionsTime Marching
01

Why plasma propulsion

Every air-breathing engine in service today — turbofan, turbojet, propeller, ramjet, scramjet — raises the thermal energy of the working fluid through combustion, converting hydrocarbon chemical energy into heat with carbon dioxide and water as byproducts. A microwave plasma jet engine proposes to reach the same end by a different route: heat ambient air electrically until it ionizes, then expand it through a nozzle. No carbon fuel, and no combustion.

This project dissects an experimental plasma jet engine built at Wuhan University to explain how it generates thrust, and builds a model of the plasma physics that makes it work. The central mechanism is collision between free electrons and neutral atmospheric particles, creating a plasma with enough thermal energy to be converted into kinetic energy through nozzle expansion.

Figure 1: Schematic of the Wuhan plasma jet engine — magnetron, air compressor, RF power supply, compressed waveguide, and quartz tube, with the plume at increasing input power.
Figure 1: Schematic of the Wuhan plasma jet engine — magnetron, air compressor, RF power supply, compressed waveguide, and quartz tube, with the plume at increasing input power.
02

Model construction

The simulation marches a 3D finite-difference grid forward in time from the moment of ignition. Each step follows the same chain: heat the electrons with the microwave field, check whether they have reached ionization threshold, ionize if so, then feed the energy exchange back into the gas.

  1. 01Initialize gas properties from the standard atmosphere and the ideal gas law.
  2. 02Compute neutral collision cross sections from atomic radii, σ = π(2r)².
  3. 03Initialize microwave properties and define the electric field as a standing wave in space and time.
  4. 04Compute electron velocity from electron temperature.
  5. 05Test electron temperature against the impact-ionization thresholds for N₂ and O₂ separately.
  6. 06Where threshold is met, compute collision frequency and update ion and neutral densities — ions rising, neutrals falling, over the timestep.
  7. 07Use plasma conductivity to compute power absorbed from the microwave field, and update temperature for the next step.
  8. 08Record and plot the spatial variation, then repeat.
03

Results

The simulation produces an animated evolution of every tracked property — electron density, neutral and ion densities for both species, electron temperature, gas temperature, and pressure — across the first 25 microseconds after ignition.

Figure 2: Plasma and gas properties at t = 0.03 µs — heating has begun but ionization has not.
Figure 2: Plasma and gas properties at t = 0.03 µs — heating has begun but ionization has not.
Figure 3: t = 0.10 µs — the core has crossed ionization threshold and electron density is rising.
Figure 3: t = 0.10 µs — the core has crossed ionization threshold and electron density is rising.
Figure 4: t = 0.17 µs — ionization spreading outward from the centre.
Figure 4: t = 0.17 µs — ionization spreading outward from the centre.
Figure 5: t = 0.25 µs — the developed state, with a clearly centre-peaked electron temperature field and the corresponding gas pressure rise.
Figure 5: t = 0.25 µs — the developed state, with a clearly centre-peaked electron temperature field and the corresponding gas pressure rise.
04

Where this leaves the concept

There is a long way to go before this architecture scales to large-aircraft propulsion, and the limiting factor is not the plasma physics — it is the electrical input required relative to the thrust generated. The mechanism works; the energy budget is the open problem.