ANNOUNCEMENT

NOBEL PRIZE 2026

This year's laureates, their fields, and the discoveries that earned them the prize.Live nowSee the winners
Moonbase
← Back to Awards
NSFDirectorate for Mathematical and Physical Sciences

Research Infrastructure: MIP: Autonomous Robotic Metallurgist: An NSF Materials Innovation Platform (ARM-MIP)

Active
Institution
Texas A&M Engineering Experiment Station, TX
Principal investigator
Raymundo Arroyave
Funding
$24.9M
Duration
2026–2032
Open portfolio

Moonbase score

Still being scored

Abstract

Nontechnical Description The Autonomous Robotic Metallurgist Materials Innovation Platform (ARM-MIP) at Texas A&M University is an open-access national user facility that designs, makes, tests, and learns from new alloys, closing the discovery loop by integrating robotics, artificial intelligence, physics-based simulation, and automated experimentation into a single autonomous laboratory. Advanced alloys underpin many nationally critical technologies, from transportation and energy systems to medical devices, yet bringing a new alloy from concept to deployment can take more than a decade. The facility funded by this award supports both experimental and computational research: scientists and engineers from universities, national laboratories, and industry come to ARM-MIP as hands-on users, running their own discovery campaigns and carrying autonomous-laboratory techniques back to their home institutions. Automation handles sample fabrication, characterization, and routine computation, while researchers focus on hypothesis generation, interpretation, and the metallurgical decisions that drive discovery. Remote access modes, including artificial-intelligence-guided campaign design and digital-twin-supported experiment planning, extend advanced discovery capabilities to additional researchers, including those at non-R1 and EPSCoR institutions. Industry engages through fee-for-service, collaborative, and consortium-based modes linking U.S. manufacturers to autonomous discovery. Training is delivered through on-site user instruction, the established Computational Materials Science Summer School (CMS3), and dedicated modules in artificial-intelligence-enabled discovery workflows. ARM-MIP fosters a national community of practitioners across academia, industry, and national laboratories who share tools, samples, data, and know-how. ARM-MIP expands U.S. capacity to translate accelerated discovery into deployable technologies. Technical Description ARM-MIP integrates high-throughput synthesis and processing, automated multimodal characterization, physics-informed simulation, and Bayesian decision-making within a single closed-loop, artificial-intelligence-orchestrated facility housed at Texas A&M University. The platform extends the closed-loop vision of the Materials Genome Initiative to autonomous, microstructure-aware bulk alloy discovery. Capabilities include automated vacuum arc melting and directed-energy deposition, high-throughput thermo-mechanical processing, automated electron microscopy with diffraction-based phase analysis, focused-ion-beam tomography, and high-throughput mechanical, thermal, and magnetic property measurement. A federated platform makes data, code, samples, and workflows findable, accessible, interoperable, and reusable. In-house research pursues two coordinated thrusts. A materials-science thrust focuses on metastable, chemically and microstructurally complex alloys that exhibit transformation-mediated functional and structural behavior. The thrust examines how composition, processing path, and microstructure jointly govern alloy properties and whether microstructure-aware autonomous discovery can reach regions of the property landscape inaccessible to conventional composition-only approaches. A complementary methodological thrust develops the autonomous-laboratory capabilities themselves: physics-informed Bayesian decision-making, agentic artificial-intelligence systems that orchestrate end-to-end campaigns and generate hypotheses, multi-modal foundation models for joint composition-processing-structure-property representation, digital-twin orchestration that validates robotic workflows virtually before physical execution, and human-robot collaboration for sample handling. ARM-MIP integrates the core elements of a Materials Innovation Platform — tool development, focused in-house research, open user access, workforce training, and broad knowledge sharing — within a single national capability for autonomous alloy discovery. This award by the Section of Materials Research in the Directorate for Mathematical and Physical Sciences is partially supported by the Section of Civil, Mechanical, and Manufacturing Innovation in the Directorate for Engineering. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Research team

  • Raymundo ArroyavePI
  • Enrique J LaverniaCo-PI
  • Ibrahim KaramanCo-PI
  • Shuiwang JiCo-PI
  • Minghui ZhengCo-PI

LOADING MOONBASE SCORE

Directorate for Mathematical and Physical Sciences01002627DB NSF RESEARCH & RELATED ACTIVIT01002930DB NSF RESEARCH & RELATED ACTIVIT01002829DB NSF RESEARCH & RELATED ACTIVIT01003031DB NSF RESEARCH & RELATED ACTIVIT01002728DB NSF RESEARCH & RELATED ACTIVIT01003132DB NSF RESEARCH & RELATED ACTIVITSpecial Initiatives, Materials Innovation Platforms, OFFICE OF MULTIDISCIPLINARY ACuniversitysystemsreviewnationallyvirtuallytechniquesinitiativecriteriameritfocuses

Questions for Texas A&M Engineering Experiment Station

Answered by the team behind this research.

Same researchers, same institution, same topic, scored first.