Seattle-based Interlune says it has managed to produce 99% pure helium-3 from a standard supply of industrial-grade helium, marking a milestone for a technology that the company aims to use on the moon.

The process, known as Cold Capture, could be profitably used on Earth even before Interlune begins lunar mining operations.

Only 0.000137% of the world’s helium exists in the form of helium-3, as opposed to the much more common helium-4 isotope. But helium-3 is uniquely suited for use as a refrigerant for quantum computers. It can also be used in radiation detectors, medical scanners and eventually fusion reactors.

Because of its rarity and utility, the price of helium-3 can range as high as $20 million per kilogram ($9 million per pound). Interlune is betting on the proposition that helium-3 is more abundant and easier to access on the moon, due to the lunar surface’s exposure to the solar wind. If Interlune’s business model works out, the company will be able to turn a profit by delivering lunar helium-3 to Earth for industrial applications.

Interlune’s first objective was to show that Cold Capture could work as advertised. The process uses cryogenic distillation to separate helium-3 from ordinary helium at temperatures approaching absolute zero.

“Capturing helium-3 from existing helium sounds deceptively simple,” Gary Lai, Interlune’s chief technology officer, said in a news release. “But helium-3 and ordinary helium are almost chemically identical, making them extraordinarily difficult to separate. Cold Capture exploits subtle physical differences between the two isotopes at cryogenic temperatures to recover helium-3 in a process designed to scale.”

Interlune demonstrated Cold Capture at a small scale in early 2025, and received a $1.25 million small-business grant from the Department of the Air Force last November to scale up the technology for commercial production.

Based on the experiments conducted since then, Interlune projects that its technology could triple the current domestic production rate of helium-3.

“Every liter of helium produced in the world contains trace amounts of helium-3,” said Rob Meyerson, co-founder and CEO of Interlune. “Cold Capture plugs into existing helium liquefaction plant infrastructure to recover that helium-3 and turn it into a valuable product.”

Interlune has already struck deals with the U.S. Department of Energy and Maybell Quantum to deliver shipments of helium-3. The first shipments are likely to come from terrestrial sources of helium, courtesy of Cold Capture.

Meanwhile, the company is following a step-by-step plan for lunar prospecting and production. A camera designed to estimate lunar levels of helium-3 is due for delivery to the moon late this year aboard Astrobotic’s Griffin-1 lander.

That mission, known as Crescent Moon, is expected to open the way for a NASA-supported experiment called Prospect Moon in 2028. The experiment will test methods to extract gases such as helium-3 and hydrogen from lunar soil and rocks.

Follow-up missions could focus on harvesting hydrogen for rocket fuel and other lunar power applications, while also collecting helium-3 for delivery to Earth.

Interlune was founded in 2020 and reported raising $18 million in seed capital in 2024. This January, the company announced an additional $5 million investment offering aimed at advancing key technical milestones.

There are some obvious big picture issues that stand between us and useful quantum computing. Issues like whether we can make enough high-quality hardware qubits to connect into the error-corrected logical qubits we need, and how we generate the states needed to perform universal computation on those logical qubits. But there are also many less prominent challenges that will need to be solved before we can perform calculations.

One of those challenges, which only affects some types of hardware, is calibration. For devices we manufacture, like superconducting qubits, there are always subtle variations among individual qubits. (This is not true when we use something like an atom to hold the qubit, but the lasers that control them can drift.) As a result, this hardware is put through a process called calibration, where we test different frequencies and amplitudes of the microwave pulses that control them to find the combination that produces the lowest error rates, and then save those settings for use in calculations.

However, you can't perform the typical calibration process while you're doing calculations, which means drift becomes an issue for long and complicated algorithms. Google, though, has figured out that it's possible to do calibration using the same data that's used for error correction.

Read full article

Comments

© Google

The massive round was co-led by ARCH Venture Partners, Spark Capital, and Khosla Ventures.