VASIMR (Variable Specific Impulse Magnetoplasma Rocket)

A plasma engine that lets you change gears: brute thrust to leave low orbit, then dial up to ultra-efficient cruise on the way to Mars.

Accelerates a fully ionised plasma — typically argon, sometimes deuterium or krypton — to exhaust velocities of 30,000–50,000 m/s, ten to fifteen times faster than the best chemical rockets. By adjusting the power split between the first-stage helicon ioniser and the second-stage ion-cyclotron heater, the operator can continuously vary the specific impulse (Isp) from roughly 3,000 s (high-thrust mode) to over 30,000 s (high-efficiency mode), all at the same electrical input power. The design target is to enable long-duration interplanetary cargo missions and possibly fast crewed Mars transits if a sufficiently powerful onboard reactor becomes available.

Stage one — the helicon coupler — wraps a quartz tube with a half-helical RF antenna driven at roughly 10–30 MHz. The standing helicon wave inside the gas-filled tube ionises argon to >99% efficiency in a fraction of a millisecond, producing a cold (~5 eV) dense plasma at moderate pressure. Stage two — the ion-cyclotron resonance heating (ICRH) stage — is a coil tuned to the ion gyrofrequency in the local axial magnetic field. As ions drift downstream into the ICRH region, they enter a moving frame in which the RF rotates with their gyration; energy is resonantly pumped into the perpendicular (rotational) motion. A diverging magnetic nozzle (typically 1–2 T at the throat, dropping to ambient at the exit) then adiabatically converts that perpendicular energy into directed axial velocity as the field lines flare outward — analogous to a de Laval nozzle but using magnetic field lines instead of physical walls. Superconducting REBCO magnets generate the 2 T field continuously without melting the structure that contains 1-million-K plasma a few centimetres away.

In plain terms

A regular rocket is like a bicycle with one gear — fast off the line, but bad at long-distance cruising. A car has gears so you can pick low-and-strong for a hill or high-and-efficient on the highway. VASIMR is the first rocket engine that has gears: turn one knob and you get strong push, turn it the other way and you get extreme fuel efficiency for the long haul.

Also called: magnetoplasma rocket, Ad Astra VX-200SS, VX-200, ICRH thruster.

Why it matters

The Ad Astra Rocket Company VX-200SS engine completed an 88-hour continuous-firing test at 80 kW in July 2021 — the longest continuous high-power electric-rocket firing on Earth at that time (operation stopped automatically at 88.1 h due to a faulty sensor).
Variable Isp lets a single engine handle both orbit-raising (where thrust matters) and interplanetary cruise (where exhaust speed matters), reducing the total mission fuel mass by 30–60% relative to a fixed-Isp Hall thruster of equivalent power.
Unlike a gridded ion engine, VASIMR has no immersed electrodes in the plasma — the only thing the multi-million-degree exhaust touches is a magnetic field. This in principle eliminates erosion-limited engine lifetime, the dominant failure mode of conventional electrostatic thrusters.
VASIMR is the textbook example of how fusion-grade plasma physics (helicons, ICRH, magnetic nozzles, REBCO superconductors) has crossed over into spacecraft propulsion research.

Where it lives

Ad Astra Rocket Company VX-200SS — ground-test engine, argon propellant, 200 kW class, achieved 88-hour continuous run at 80 kW in July 2021 (Webster, TX)
Ad Astra VX-200 (earlier) — vacuum chamber prototype demonstrating 70% efficiency in ICRH stage (Acta Astronautica, 2009)
NASA NextSTEP partnership — provided continuing support 2015–present for VASIMR ground-testing toward potential lunar Gateway demonstration
ESA HiPER concept — European 100 kW class helicon-plus-ICRH propulsion study citing VASIMR architecture (2010 ESA contract report)

Key numbers

Power class (VX-200SS): 200 kW (target); 80 kW (demonstrated 88-hour) (Future flight units sized for 100–200 kW solar-electric or nuclear-electric platforms)
Exhaust velocity: 30,000–50,000 m/s (Equivalent to Isp 3000–5000 s in high-thrust mode; >30,000 s in efficiency mode)
Plasma temperature (ICRH exit): ~10⁶ K (Ion energy ~50–100 eV; electrons cooler)
Magnetic field at throat: ~2 T (REBCO HTS coils; field falls off to ambient in <1 m for adiabatic expansion)
Propellant: Argon (primary); deuterium / krypton studied (Argon is cheap, inert, easy to ionise, and stores at moderate pressure)
Ionisation efficiency (helicon): >90% at 30 kW input (Acta Astronautica review, Squire & Chang Díaz 2009)
Continuous-run record: 88 h at 80 kW (VX-200SS, July 2021) (Ad Astra AIAA ASCEND 2021 paper (Cassady et al.))

Surprising facts

The exhaust plasma reaches roughly 1 million K in the ICRH stage — hotter than the Sun's photosphere. It is contained entirely by magnetic field; nothing physical touches it on the way out.
VASIMR was conceived in 1979 by Costa Rican-American astronaut Franklin Chang Díaz while a graduate student at MIT, decades before he flew seven Space Shuttle missions — a record-tying number of human spaceflights.
Specific impulse can be changed mid-firing without changing hardware. The same engine, on the same propellant tank, runs at 3,000 s during orbit-raising and at 30,000 s during cruise — physically the same atoms come out a thousand times faster in cruise mode.
The VX-200SS magnetic nozzle uses high-temperature superconducting REBCO tape magnets at ~50 K, the same family of conductors now powering Commonwealth Fusion Systems' SPARC tokamak — direct cross-fertilisation between fusion and propulsion research.

Open questions

Magnetic-nozzle detachment problem: For VASIMR to produce thrust, the plasma must eventually detach from the diverging magnetic field lines and continue as a free supersonic jet — otherwise the magnetic 'tail' would loop back and grab the spacecraft. The exact detachment mechanism (super-Alfvénic ion inertia, plasma instability, ambipolar electric field, magnetic reconnection) is still actively debated in the plasma-physics literature. Ground tests in vacuum chambers show detachment occurs, but the mechanism dominant in deep-space conditions has not been directly measured.

A short history

Franklin Chang Díaz conceived the variable-Isp magnetoplasma rocket as a graduate student at MIT in 1979, building on tokamak-fusion research into helicon and ion-cyclotron heating of magnetised plasmas. He continued the work at the Charles Stark Draper Laboratory and later at the NASA Advanced Space Propulsion Laboratory inside Johnson Space Center, where the first VASIMR prototypes were built in the 1990s. In 2005 NASA spun the program out to the Ad Astra Rocket Company, founded by Chang Díaz, which built the VX-200 (operational 2009) and VX-200SS (operational 2014, 88-hour run July 2021) ground-test engines. The 2015 NASA NextSTEP-1 award funded long-duration ground qualification toward a possible flight demonstration on the lunar Gateway in the late 2020s. As of 2026 no VASIMR engine has flown in space, but ground testing continues at progressively higher continuous power levels.

Sources

Squire, J.P. & Chang Díaz, F.R. — Acta Astronautica 64, 526 (2009): VASIMR engine performance results (https://doi.org/10.1016/j.actaastro.2008.07.013)
Ad Astra Rocket Company — Technical Library and engineering reports (https://www.adastrarocket.com/aarc/Technology)
NASA NextSTEP program — Advanced Electric Propulsion Systems (https://www.nasa.gov/directorates/spacetech/nextstep/)
Chang Díaz, F.R. — Scientific American, November 2000: 'The VASIMR Rocket' (https://www.scientificamerican.com/article/the-vasimr-rocket/)
Cassady et al. — IEPC 2017 paper: VX-200SS Engineering Demonstration Unit (https://www.adastrarocket.com/Cassady_IEPC2017_VX200SS.pdf)