English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We introduce GHP-MOFassemble, a Generative artificial intelligence (AI), HighPerformance framework to accelerate the rational design of metal-organicframeworks (MOFs) with high CO2 capacity and synthesizable linkers. Ourframework combines a diffusion model, a class of generative AI, to generatenovel linkers that are assembled with one of three pre-selected nodes into MOFsin a primitive cubic (pcu) topology. The CO2 capacities of these AI-generatedMOFs are predicted using a modified version of the crystal graph convolutionalneural network model. We then use the LAMMPS code to perform molecular dynamicssimulations to relax the AI-generated MOF structures, and identify those thatconverge to stable structures, and maintain their porous properties throughoutthe simulations. Among 120,000 pcu MOF candidates generated by theGHP-MOFassemble framework, with three distinct metal nodes (Cu paddlewheel, Znpaddlewheel, Zn tetramer), a total of 102 structures completed moleculardynamics simulations at 1 bar with predicted CO2 capacity higher than 2 mmol/gat 0.1 bar, which corresponds to the top 5% of hMOFs in the hypothetical MOF(hMOF) dataset in the MOFX-DB database. Among these candidates, 18 have changein density lower than 1% during molecular dynamics simulations, indicatingtheir stability. We also found that the top five GHP-MOFassemble's MOFstructures have CO2 capacities higher than 96.9% of hMOF structures. This newapproach combines generative AI, graph modeling, large-scale molecular dynamicssimulations, and extreme scale computing to open up new pathways for theaccelerated discovery of novel MOF structures at scale.

GHP-MOFassemble: Diffusion modeling, high throughput screening, and  molecular dynamics for rational discovery of novel metal-organic frameworks  for carbon capture at scale