AI Searched 100 Million Possibilities and Found a Cheaper Way to 3D-Print a NASA Rocket Alloy

Washington State University researchers used AI to navigate over 100 million possible 3D-printing configurations for NASA's GRCop-42 alloy — a copper-chromium-niobium material built for extreme heat in rocket engines. After just 40 experiments, the system found six working setups, including a record-low 500-watt configuration. This could democratize access to the alloy by enabling printing on widely available commercial machines instead of expensive high-power systems.

Key Takeaways

Washington State University researchers used AI to navigate over 100 million possible 3D-printing configurations for NASA's GRCop-42 alloy — a copper-chromium-niobium material built for extreme heat in rocket engines.
After just 40 experiments, the system found six working setups, including a record-low 500-watt configuration.
This could democratize access to the alloy by enabling printing on widely available commercial machines instead of expensive high-power systems.

AI synthesis3 sources6

Washington State University researchers have used artificial intelligence to identify a faster and less costly way to 3D print a high-performance metal alloy developed by NASA, avoiding the need to manually test more than 100 million possible printing configurations.

The material, GRCop-42, is an alloy made from copper, chromium, and niobium. NASA developed it for demanding environments where both heat resistance and efficient heat transfer are essential — particularly liquid rocket engine combustion chambers. Despite its desirable properties, the alloy is difficult and costly to 3D print because the process typically requires substantial laser power and energy. Ninety percent of commercial 3D printers cannot currently print it.

Previous attempts to print GRCop-42 using the lower wattages available on more common commercial machines had not succeeded. Testing possible printing settings one by one is impractical: each attempt consumes expensive material, requires specialized equipment, and takes considerable time.

The researchers started with data from 37 printing configurations that had already failed in earlier experiments. Using those results, they developed an AI method that could estimate how likely an untested combination of settings was to produce a successful print. The model then recommended small groups of new configurations to test, balancing exploratory and exploitative choices.

Despite the odds — successful settings were extremely rare among the 100 million possibilities — the team found six successful configurations at different laser power levels during three months of work, limiting the project to a total of just 40 experiments. For the first time, they successfully printed GRCop-42 using just 500 watts of laser power.

The work, published in the Proceedings of the AAAI Conference on Artificial Intelligence and awarded the Innovative Deployed Application Award, could eventually make the alloy printable on much more widely available equipment, opening access to universities, smaller laboratories, and companies without specialized high-power printing systems. The researchers say the same AI-guided approach could be adapted for other metal alloys and has potential applications in drug discovery and other fields where successful results are rare among enormous search spaces.