Propulsion Methods Comparison
Every propulsion method humans have flown can be ranked by one brutal number: years to Proxima Centauri. The fastest proven technology takes 425 years. The fastest credible plan in development takes 21 years — but cannot slow down when it arrives. The technologies that could decelerate and carry humans don't exist yet, and some may never.
This page lines them up by that number, and by the harder one: can you stop.
The lineup
| Technology | Peak speed | Specific impulse | Years to Proxima (4.24 ly) | Years to TRAPPIST-1 (39 ly) | Status |
|---|---|---|---|---|---|
| tech ion drive | ~0.01c | 3,000–12,000 s | ~425 | ~3,900 | Flown (Dawn, Hayabusa) |
| tech solar sail | 0.01–0.1c | No propellant | 42–425 | 390–3,900 | Flown (IKAROS 2010, LightSail 2) |
| tech laser propulsion | 0.2c | No propellant | ~21 | ~195 | Concept (mission breakthrough starshot) |
| tech nuclear pulse | 0.03–0.05c | 10,000–100,000 s | 85–140 | 780–1,300 | Tested at small scale, treaty-banned |
| tech fusion drive | 0.05–0.12c | 100,000–1,000,000 s | 35–85 | 325–780 | No working engine, no working reactor |
| tech antimatter drive | 0.5–0.9c | ~10,000,000 s | 5–8 | 43–78 | Production: ~nanograms/year |
| tech alcubierre drive | Effective >c | — | Minutes? | Hours? | Needs negative-energy matter |
The four tiers
Tier 1 — Flown. Ion drives (NASA's Dawn ran on xenon to Vesta and Ceres; Hayabusa-2 returned asteroid samples; Starlink uses Hall thrusters) and solar sails (JAXA's IKAROS deployed a 14m sail near Venus in 2010; LightSail 2 raised its orbit on sunlight in 2019). Both work. Neither is fast enough to matter for interstellar distances — solar pressure falls as 1/r², so a sail's window of useful thrust closes inside Saturn's orbit.
Tier 2 — In development. tech laser propulsion, the only near-term interstellar plan with a credible budget. Breakthrough Starshot's design: a ~100 GW phased-array laser pushes gram-scale sails to 0.2c over a few minutes. The acceleration is roughly 10,000 g. The known unsolved pieces: phased-array coherence at scale, sail material that survives the beam, and — the unsolved one — how to decelerate at the destination. Current answer: you don't. You fly through the Proxima system in 4 minutes and hope your sensors caught something.
Tier 3 — Theoretical, physics-OK. mission project orion (1958–1965) studied detonating shaped nuclear charges behind a pusher plate. Test stands at General Atomics showed the principle worked at chemical-explosive scale. The 1963 Partial Test Ban Treaty ended it. Fusion drives — VASIMR, Direct Fusion Drive, Project Icarus — assume sustained fusion in a flight-weight engine. We don't have sustained fusion in a building-sized one.
Tier 4 — Speculative. Antimatter annihilation gives the theoretical maximum energy per mass (E=mc², 100% conversion). CERN and Fermilab combined produce roughly nanograms of antiprotons per year at billions of dollars per gram. No one has held antihydrogen for longer than minutes; engine-relevant quantities are not on any roadmap. The tech alcubierre drive (Miguel Alcubierre, 1994) proposed a metric that contracts space ahead and expands it behind, letting a ship inside the bubble move faster than light without locally violating relativity. The original solution required negative energy on the order of a Jupiter mass. Harold White's 2011 refinement dropped this by orders of magnitude. Both still require exotic matter — no one has produced any.
The deceleration problem
This is the trade-off table that matters:
| Factor | Ion / chemical | Nuclear / fusion | Laser sail | Antimatter | Warp |
|---|---|---|---|---|---|
| Carries propellant | Yes | Yes | No | Yes | N/A |
| Can decelerate | Yes | Yes | No | Yes | Yes |
| Two-way trip | Yes | Yes | Flyby only | Yes | Yes |
| Where infrastructure lives | On the ship | On the ship | Earth-side (100 GW array) | Production facility | Unknown |
| Payload | Tons | Tons | Grams | Tons | Unknown |
Every propellant-based drive pays a rocket-equation tax: to slow down at the destination, you carry the fuel to do it, which you accelerated, which required more fuel. The math is exponential. A laser sail dodges this by leaving its engine on Earth — but then has no engine when it arrives. Proposed answers (magnetic braking against the interstellar medium, photon braking off the destination star's light) have been sketched but never tested at any scale.
What's contested
Is the Alcubierre drive physically possible? Quantum field theory permits negative energy density in small regions (Casimir effect, squeezed vacuum states). Whether macroscopic, persistent negative-energy matter can exist is open. The 2021 Lentz and Bobrick-Martire papers proposed warp geometries using positive-energy matter; both remain contested.
Can fusion drives actually scale down? ITER aims to demonstrate sustained net-positive fusion in a 23,000-tonne reactor. A flight engine has to do similar physics in maybe 100 tonnes. No clear path.
Is laser-sail deceleration solvable? The honest answer is nobody knows. If it isn't, then 0.2c flybys are the ceiling, and "visiting" Proxima means a few minutes of data acquisition.
Why this has to do with other realms
The propulsion table is, underneath, an concept energy density table. Chemical fuels: ~10 MJ/kg. Nuclear fission: ~80 TJ/kg. Fusion: ~340 TJ/kg. Antimatter annihilation: ~90 PJ/kg — the ceiling set by E=mc². The reason interstellar travel is hard is that even at 100% efficiency, the mass-energy of the propellant you'd need approaches the mass of the ship.
This is also a concept fermi paradox argument. If interstellar travel is locked behind energy densities only antimatter or warp can supply, then "where is everybody" has a boring answer: it's expensive, and even advanced civilizations might decide it isn't worth it. Compare with concept kardashev scale — only a Type II civilization has the energy budget to send mass between stars at relativistic speeds at any frequency.
An open question
If the only interstellar technology that works in human-relevant timescales is a gram-scale flyby that cannot stop, what does it mean to say humanity has "reached" another star? At what point does sending instruments stop being exploration and start being concept deep time messaging — a message in a bottle thrown toward a coastline you will never see?
Key sources
- Project Orion: The True Story of the Atomic Spaceship by George Dyson (2002) — load-bearing reference for the nuclear-pulse history.
- Alcubierre, M. (1994) "The warp drive: hyper-fast travel within general relativity," Classical and Quantum Gravity 11 — the original metric.
- Breakthrough Starshot technical documentation at breakthroughinitiatives.org — the 100 GW / 0.2c numbers come from here.
- To verify: Lentz (2021) and Bobrick-Martire (2021) positive-energy warp solutions — recent, contested, worth tracking.
- Frontiers of Propulsion Science (Millis & Davis, eds., 2009, AIAA) — the closest thing to a textbook on this terrain.
Further reading
- mission breakthrough starshot — the engineering targets in detail, and why they're harder than the press releases say.
- mission voyager 1 — what 0.005% of the way to the nearest star looks like after 47 years.
- The Starflight Handbook by Mallove & Matloff (1989) — dated but still the cleanest first-principles treatment.
- Isaac Arthur's "Upward Bound" YouTube series — speculative but mathematically careful.
- To verify: Centauri Dreams blog (Paul Gilster) — running commentary on interstellar propulsion research since 2004.
See Also
- compare travel times (the same numbers, organized by destination instead of by drive)
- overview distance scales (why 4.24 light-years is the easy case)
- concept rocket equation (the exponential tax every propellant-based drive pays)
- concept energy density (propulsion is energy density, cross-realm into physics and engineering)
- concept fermi paradox (if interstellar is this hard, the silence has a mundane explanation)
- tech generation ship (the architecture that gives up on speed and trades it for time)