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

What happens when you combine artificial intelligence with metal 3D printing? Researchers at Washington State University (WSU) just demonstrated the answer: a breakthrough that could democratize the production of one of NASA’s most demanding alloys.

Their AI system sifted through more than 100 million possible printing configurations for GRCop-42, a high-performance copper alloy developed by NASA for rocket engines. After just 40 experiments, the system identified six successful settings — including the first-ever successful print at only 500 watts of laser power.

What Is GRCop-42 and Why Is It So Hard to Print?

GRCop-42 is a copper-chromium-niobium alloy engineered by NASA for environments that demand both extreme heat resistance and efficient heat transfer. It is widely used in liquid rocket engine combustion chambers, where temperatures can soar to thousands of degrees while the material must still conduct heat effectively to prevent engine failure.

The problem? 3D printing GRCop-42 traditionally requires enormous laser power — far beyond what most commercial metal printers can deliver. Previous attempts to print the alloy at lower wattages on standard equipment simply failed. The material would overheat, warp, or melt entirely.

“Sometimes they printed a certain configuration, and the product just melted,” said Azza Fadhel, a PhD student in computer science at WSU and first author of the study. “It wasn’t really printable, and even with time and money, they wouldn’t be able to try all 100 million options.”

Each test print costs hundreds of dollars in materials alone, and detailed quality analysis of a finished sample can take days. Manually testing all possible combinations of laser power, scan speed, layer thickness, and other parameters was simply impractical.

How AI Cracked the Code in Just 40 Experiments

The WSU team — spanning the School of Electrical Engineering and Computer Science and the School of Mechanical and Materials Engineering — took a different approach. Instead of brute-force testing, they trained an AI model on data from just 37 previously failed experiments.

Using those failures as learning material, the system developed a method to estimate how likely any untested combination of printing parameters was to succeed. The AI then recommended small batches of new configurations that balanced two goals: testing the most promising options and exploring uncertain areas that could improve the model’s predictions.

“They would give me back the results, and I liked all of them — even if they failed — because every result improved our AI model,” Fadhel explained.

Over three months, the team conducted just 40 total experiments and found six successful configurations at different laser power levels. The standout result: the first successful 3D print of GRCop-42 using only 500 watts of laser power — a level available on many commercial metal printers.

What This Means for 3D Printing and Aerospace

The implications are significant. By lowering the power threshold from specialized industrial systems to commercially available equipment, this breakthrough could:

• Democratize access — Universities, small labs, and mid-size manufacturers could print GRCop-42 without investing in million-dollar high-power systems
• Reduce costs — Lower laser power means less energy consumption, reduced equipment wear, and cheaper post-processing
• Accelerate innovation — More researchers and companies experimenting with the alloy could lead to new applications beyond aerospace

Professor Jana Doppa, who led the research, put it in perspective: “Ninety percent of commercial printers cannot print this metal alloy, so given that we were able to find these feasible process parameters, it allows us to use those commercial printers, and we are essentially democratizing the printing of this alloy.”

Beyond Rockets: Broader Applications

The research team also noted a promising application in jet engine manufacturing. Using a technique called Directed Energy Deposition (DED), GRCop-42 could be used to coat existing jet engine alloys. Even a 50-degree Celsius increase in engine operating temperature can improve fuel efficiency by up to 10 percent — a massive gain for the aviation industry.

Perhaps most exciting is the AI methodology itself. The approach — using machine learning to navigate enormous search spaces where successful outcomes are rare — could be applied to challenges far beyond 3D printing, including drug discovery and materials science.

“There is risk that we are deploying something where real people, materials, and physical costs are involved,” said Doppa. “We didn’t know whether we would succeed or not.”

The team’s work, published in the Proceedings of the AAAI Conference on Artificial Intelligence, received the Innovative Deployed Application Award at the organization’s annual conference.

How TT3DPrint Can Help

At TT3DPrint, we specialize in custom FDM 3D printing for figurines, prototypes, educational tools, and creative products. While metal alloy printing requires industrial-grade equipment, the same principle drives our work: leveraging the latest 3D printing technology to deliver precision, quality, and value.

Whether you need a one-off prototype or a production run of custom items, our 220+ printer fleet is ready to bring your designs to life. Contact us for a free quote.

Conclusion

Washington State University’s research demonstrates a powerful new paradigm: using AI to dramatically accelerate the discovery of viable 3D printing parameters. By reducing 100 million possibilities down to 40 targeted experiments, the team has opened the door to printing NASA-grade alloys on commercial equipment — a development that could reshape aerospace manufacturing and inspire similar breakthroughs across industries.