WSU Researchers Used AI to Make a NASA Alloy 3D Printable on Regular Machines

What Is GRCop-42 and Why Does It Matter

GRCop-42 is a copper-chromium-niobium alloy NASA developed for the inside of liquid rocket engine combustion chambers. It handles extreme heat while staying thermally conductive, which makes it one of the most sought-after materials in aerospace additive manufacturing. The problem: commercial 3D printers almost never have the laser power to process it reliably. That puts it out of reach for most universities, small labs, and mid-sized manufacturers.

The AI Approach

Researchers at Washington State University's School of Electrical Engineering and Computer Science and School of Mechanical and Materials Engineering took a different approach. Instead of trying every laser setting by hand, they fed 37 previously failed print configurations into an AI model. The model estimated which untested settings were most likely to succeed and recommended small batches to test.

Over three months and 40 total experiments, the team identified six successful configurations. One worked at 500 watts, a first for this alloy on commercial-class equipment.

Why 500 Watts Is the Big Number

GRCop-42 typically demands much higher laser power than most commercially available LPBF systems can deliver. Getting it to print at 500 watts means labs with standard machines no longer need to rent or buy specialized high-power systems just to experiment with the material. Print costs drop, equipment wear decreases, and the barrier to entry shrinks significantly.

Broader Impact

Jana Doppa, who led the research, put it plainly: 90% of commercial printers cannot print this alloy. Finding feasible process parameters lets researchers use those machines and democratize access to GRCop-42.

The AI framework itself is reusable. The same approach could speed up process discovery for other metal alloys, and the underlying method applies to other fields where experiments are costly and the search space is huge, including drug discovery. The team published their findings in the Proceedings of the AAAI Conference on Artificial Intelligence and received the Innovative Deployed Application Award.