Traditional materials development is slow, costly, and limited to exploring only a tiny fraction of possible compositions. Yet the chemical design space for RCCAs contains billions of potential alloys. RADISH bridges this gap by combining computational modeling, combinatorial synthesis, high-throughput characterization, and machine learning into one powerful discovery pipeline.

Our workflow integrates:

• High-throughput CALPHAD simulations to scan vast compositional spaces and pinpoint the most promising alloy regions.

• Combinatorial material libraries, produced in partnership with leading research institutions, enabling hundreds of compositions to be synthesized in a single experiment.

• High-throughput characterization (XRF, XRD, ERDA, nanoindentation, micromechanics) to rapidly map structure–property relationships.

• AI-powered prediction models that learn from experimental data and forecast the behavior of unexplored alloys.

• Upscaling and high-temperature testing of the top candidates, including spark plasma sintering, phase stability analysis, creep testing, and oxidation resistance at extreme temperatures.

The goal is clear: to discover RCCAs that combine high strength at 1000°C, ductility at room temperature, exceptional oxidation resistance, and reduced density-performance levels unattainable with current engineering materials.