The Elusive Aerospike Engine Is Finally Moving From Theory to Reality

Here’s what you’ll learn when you read this story:

The aerospike engine—which uses air pressure to shape the downward force needed to launch a spacecraft—has been on the cusp of entering mainstream rocketry for decades.
Now, a new company called Leap 71 is moving fast, designing a new aerospike engine and 3D printing it for testing all within a few weeks.
The successful test this past December paved the way for the company’s adoption of a reference design called the Noyron XRA-2E5, which could be tested no later than the end of 2026.

For 65 years of human spaceflight, rockets ferrying astronauts to the stars have all had bell-shaped nozzles specifically designed to produce enough thrust to lift massive payloads off the launchpad. But this decades-old design comes with limitations.

The biggest one is that bell nozzles only work optimally at certain atmospheric pressures, which is one reason why most rockets actually contain many stages during a mission. If you could improve this inefficiency, rockets could just have one single stage that they use for the entire mission. That would save on payload weight and, of course, money.

So, for decades, scientists have wondered whether other geometric designs—ones that could effectively adjust to atmospheric pressure during flight—could one day replace bell-shaped nozzles. The leading contender is an idea known as the aerospike, which relies on atmospheric pressure itself to create the outside wall of a virtual bell. The idea is that as atmospheric pressure changes, the bell itself would change as well. The idea isn’t new, as this rocket engine—designed in both toroidal and linear configurations—has been on the precipice of the mainstream ever since NASA strapped one to an SR-71 in the 90s.

However, the engine has never proven itself to be adequate replacement for traditional rocket bells. But times have changed, and in 2024, the aerospike had one of its best years in recent memory. In late October, the German aerospace company Polaris Spaceplanes successfully demonstrated a linear aerospike (LAS) engine mid-flight. Now, the company hopes to launch a successor, called the Nova, in 2025.

But it was actually a late entry from the Dubai-based computational engineering company Leap 71 that gave a glimpse of what the future research into aerospikes—and aerospace propulsion more broadly—might look like. On December 18, 2024, the company test fired its oxygen-and-kerosene-burning toroidal aerospike rocket for 11 seconds, recording a 1,110 pounds of thrust. And while that’s cool in and of itself, it’s not as impressive as how it was made.

Leap 71 used an in-house AI computation engineering model known as Noyron to design the rocket and then used a 3D-printing technique known as “Laser Powder Bed Fusion” to create an engine built from an aerospace copper alloy. Building a new rocket engine usually takes several years (at least), but Leap 71 claims that their aerospike engine took only a matter of weeks to design, manufacture, and test. The company even addressed one of the aerospike’s biggest design challenges—how to cool the spike itself.

“We were able to extend Noyron’s physics to deal with the unique complexity of this engine type,” Leap 71 CEO Josefine Lissner said in a press statement. “The spike is cooled by intricate cooling channels flooded by cryogenic oxygen, whereas the outside of the chamber is cooled by the kerosene fuel.”

In April 2025, the company announced that they’re moving forward with an aerospike reference design called the Noyron XRA-2E5, and it hopes to perform first tests on the design by the end of 2026.

Aerospike engines have long been waiting for a chance to prove their usefulness, and thanks to AI computational tools and rapid prototyping via 3D printing, that chance may come much sooner than anyone expected.

You Might Also Like