Nuclear Fusion Rockets Are No Longer Science Fiction and a Decade Could Prove It

A Historic Spark in a Lab in England

For most of human history, the idea of traveling to Mars in a matter of weeks existed only on the pages of science fiction novels. That may be changing. On March 26, 2026, Bletchley-based Pulsar Fusion achieved the first-ever plasma ignition inside a nuclear fusion rocket — a milestone no other company or government had previously reached. The test was not performed in secret. The achievement was demonstrated live during a dedicated technical session at Amazon's MARS Conference, hosted by Jeff Bezos, in Ojai, California, where visionary leaders in machine learning, automation, robotics, and space gathered to shape humanity's future beyond Earth.

The test used krypton gas to generate and confine superheated plasma within the company's Sunbird nuclear fusion exhaust system, using a combination of electric and magnetic fields. By showing that the system can generate and confine plasma, Pulsar Fusion validated the core technology for its Dual Direct Fusion Drive (DDFD) engine — a system the company claims will have a remarkably high specific impulse of 10,000 to 15,000 seconds. For context, chemical rockets generate extremely high thrust, essential for launch and rapid maneuvers, but their relatively low exhaust velocities limit how fast spacecraft can ultimately travel through space.

The Race to Build a Fusion Engine

Pulsar Fusion is not alone in this pursuit. U.K. startup Pulsar Fusion, California-based Helicity Space, and fusion researchers at Princeton are converging on practical fusion propulsion — each taking a different technical path toward the same audacious goal. Helicity Space closed a $5 million seed round in late 2023 with investors including Airbus Ventures, and later secured strategic capital from Lockheed Martin's venture arm to push its lab prototypes from two to four plasma guns and toward a full proof-of-concept drive. NASA's Innovative Advanced Concepts program has also funded studies of a "Helicity Drive"–powered constellation to explore the heliosphere, signaling institutional interest in using fusion propulsion for long-range science.

Fusion-powered space travel has long held the promise of rapid trips across the solar system: Mars in weeks, Saturn in months, Pluto in years. Producing meaningful thrust requires staggering numbers of fusion reactions every second — a quintillion reactions would provide about 10 newtons of thrust, equivalent to the weight of a 1-liter bottle of water in your hand. But a fusion-powered engine would provide this thrust for months, rather than minutes for typical chemical propulsion, allowing large speeds to be reached eventually.

What Sunbird Could Actually Look Like in Space

Unlike current rockets that launch from Earth, these fusion rockets would be stored on giant orbital docking stations. In a concept video, a Sunbird undocks from its station and uses its eight thrusters to attach to a larger spacecraft and propel it to a distant planet — kind of like a jet pack, but for spaceships. Once it reaches its destination, Sunbird detaches and docks with an awaiting station, allowing the rocket to repeatedly carry spacecraft to and from deep space. The DDFD is designed to deliver a specific impulse of 10,000–15,000 seconds and 2 MW of power, offering roughly 1,000 times the thrust of conventional in-orbit propulsion. Pulsar Fusion plans a 2027 in-orbit demonstration of Sunbird's core components.

If any of these efforts prove successful, missions across the solar system for robots and humans could be unlocked like never before, turning us into a true spacefaring species. The stakes are enormous, and the optimism among those building these systems is hard to miss. "If we continue on the current trajectory, everything we know about space travel is going to change within a decade," Stephane Lintner, CEO and co-founder of Helicity Space, told Live Science.

Breakthrough or Overpromise?

Skeptics are right to temper expectations. Achieving first plasma is a long way from a functioning interplanetary engine, and the history of fusion energy is littered with timelines that slipped by decades. Nuclear fusion is a process whereby two atoms combine, or fuse, to form a heavier one, releasing excess energy in the form of light and heat — the best-known type being the process that occurs inside the sun at up to 27 million degrees Fahrenheit (15 million degrees Celsius), where atoms of hydrogen fuse to form helium. Replicating and controlling that process reliably enough to power a spacecraft remains one of the hardest engineering problems ever attempted.

Still, what makes this moment different from previous decades of fusion promises is the convergence of private capital, advanced materials, and computational power — all arriving at once. The next development phase will see Pulsar gather detailed performance data, including thrust and exhaust velocity measurements, using specialized instruments, and this data will enable the company to plan the first Sunbird mission.