Structure → synthesis condition learning

• Transformer-based text mining of synthesis procedures from literature
• One-to-one mapping between curated MOF structures and experimental recipes
• Graph neural networks trained on 3D crystal structures
• Reduction of manual literature search for experimental validation

Large-scale MOF curation and automation

• Curation of 40,000+ metal–organic frameworks
• Automated correction of structures and geometry optimisation
• MOF decomposition, topology classification, porosity analysis
• Integration into the NOMAD Porosity Normaliser

Production-grade tools and workflows

• Python platforms for structure analysis and descriptor computation
• High-throughput screening workflows on HPC systems
• ASE, Pymatgen, RDKit integration
• Containerised and CI/CD-enabled research pipelines

Electronic structure & ion–framework interactions

• DFT-based electronic structure and PDOS analysis
• Identification of redox-active sites
• Framework–ion interaction studies for batteries
• Structure–property relationships for electrochemical stability

Chemistry-controlled transport and selectivity

• Effect of isomerism on gas uptake and diffusion
• Topology-controlled comparative studies
• Design rules for rational MOF functionalisation
• Direct feedback to synthesis strategies

Toward self-driving materials laboratories

• Integration of AI models, databases, and synthesis knowledge
• Robotic synthesis workflows for porous materials
• Closed-loop discovery and optimisation
• Startup-ready digital chemistry platforms