[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86495-en":3,"doc-seo-86495-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},86495,13056703019404,"Miles","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","Large language model agents accelerate inverse design of metal-organic frameworks for gas separation","Metal-organic frameworks (MOFs) enable highly modular adsorptive gas separation, but inverse design is hard because reticular spaces must satisfy chemical validity, separation quality, and structural diversity at once. LEMO Agent is a large-language-model agent framework for closed-loop inverse design in MOFid space, combining language-driven candidate generation, MOFid standardization, validity checking, Transformer-based property prediction, design memory, and multi-island exploration. Iterative generate–validate–evaluate–remember cycles use feedback from successes and failures to search across linker, metal, and topology. Experiments on CH4/N2 and CO2/N2 yield enriched high-performing candidates with broad chemical and topological diversity, validated via reconstruction, GCMC simulation, and experimental down-selection through feasibility and ligand purchasability, enabling initial wet-lab synthesis and SEM characterization.","arXiv :2607 . 10559v 1 [ cs .AI] 12 Jul 2026  \nLarge language model agents accelerate inverse design of metal–organic frameworks for gas separation  \nZhaolin Hu 1 , Hehe Fan 1,3 , Wangyihan Guo 1 , Meng Xu2 , Chenhao Rao2 ,  \nQiwei Yang2 , Yi Yang1,3  \n1 College of Artificial Intelligence, Zhejiang University , Hangzhou, China.  \n2 College of Chemical and Biological Engineering, Zhejiang University Hangzhou, China.  \n3 Institute of Fundamental and Transdisciplinary Research, Zhejiang University Hangzhou, China.  \nAbstract  \nMetal-organic frameworks (MOFs) offer a highly modular platform for adsorptive gas separation, yet their vast reticular design space makes inverse design difficult under simultaneous constraints of chemical validity, separation performance, and structural diversity. Here, we present LEMO Agent, a large-language-model agent framework for closed-loop inverse design of gas-separation MOFs in MOFid space. LEMO Agent couples language-based candidate generation with MOFid standardization, explicit validity checking, Transformer-based property prediction, structured design memory, and multi-island exploration. Through iterative generate–validate–evaluate–remember cycles, the agent uses feedback from both successful and failed candidates to guide chemically constrained search across linker, metal, and topology choices. We evaluate LEMO Agent on CH4 /N2 and CO2 /N2 separation tasks. Compared with representative generative, optimization, and agentic baselines, LEMO Agent enriches high-performing candidates, improves predicted separation performance, and maintains broad chemical and topological diversity. Selected candidates are further reconstructed, evaluated by GCMC simulations, and passed through an experimental down-selection workflow based on chemical feasibility and ligand purchasability, leading to initial wet-lab synthesis and SEM characterization. These results demonstrate that large language model agents can serve as interpretable and scalable design engines for accelerating MOF discovery beyond conventional fixed-library screening.  \nKeywords: Metal-Organic Frameworks, Large Language Models, Agent, Inverse Design, Gas Separation  \n1 Introduction  \nMetal–organic frameworks (MOFs) are crystalline porous solids assembled from metal ions or clusters and multitopic organic linkers through coordination-driven self-assembly and reticular chemistry [1–4] . The modular combination of metal nodes and organic linkers enables systematic control over pore size, pore shape, framework topology, surface functionality, and local adsorption environments, giving rise to one of the most chemically diverse families of porous materials [1, 3 , 4] . Owing to their high internal surface areas, ordered pore networks, and tunable host–guest interactions, MOFs have been widely explored for gas storage, adsorption, separation, catalysis, sensing, and related energy and environmental applications [1, 2 , 5–8] . Among these applications, adsorptive gas separation is particularly sensitive to the molecular-level matching between framework pores and gas molecules. In this context, separation performance is not governed by a single descriptor, but by the coupled effects of pore aperture, pore geometry, framework polarity, open metal sites, linker functional groups, framework flexibility, and the spatial distribution of adsorption sites [5–12] . These characteristics make MOFs an attractive platform  \nfor separation-oriented materials design, while also creating a complex chemical and structural search space in which high-performing candidates are difficult to identify by intuition or trial-and-error alone.  \nThe same structural tunability that makes MOFs attractive for separation also creates a formidable combinatorial design problem. The possible combinations of metal nodes, organic linkers, functional groups, network topologies, and interpenetration patterns far exceed the number of MOFs that have been experimentally synthesized or comp","cbCaipm2bOwbTQMF","https://ap.wps.com/l/cbCaipm2bOwbTQMF","pdf",9870662,3,1,19,"English","en",105,"# Abstract\n# 1 Introduction","[{\"question\":\"Why is inverse design of MOFs for gas separation challenging?\",\"answer\":\"MOFs must satisfy simultaneous constraints: chemical validity, separation performance, and structural diversity, within a very large reticular design space.\"},{\"question\":\"What is LEMO Agent and how does it perform inverse design?\",\"answer\":\"LEMO Agent is a closed-loop agent framework that generates MOF candidates in MOFid space, standardizes them, checks validity, predicts properties with a Transformer model, stores structured memory, and explores via multi-island search using iterative generate–validate–evaluate–remember cycles.\"},{\"question\":\"How was LEMO Agent evaluated and validated experimentally?\",\"answer\":\"It was evaluated on CH4/N2 and CO2/N2 separation tasks, compared to generative/optimization/agent baselines, reconstructed and assessed using GCMC simulations, then down-selected by chemical feasibility and ligand purchasability for initial wet-lab synthesis and SEM characterization.\"}]",1784212173,48,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"large-language-model-agents-accelerate-inverse-design-of-metal-organic-frameworks-for-gas-separation","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":20},"https://docshare.wps.com/document/research-report/",{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/large-language-model-agents-accelerate-inverse-design-of-metal-organic-frameworks-for-gas-separation/86495/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-24","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is inverse design of MOFs for gas separation challenging?","Question",{"text":75,"@type":76},"MOFs must satisfy simultaneous constraints: chemical validity, separation performance, and structural diversity, within a very large reticular design space.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What is LEMO Agent and how does it perform inverse design?",{"text":80,"@type":76},"LEMO Agent is a closed-loop agent framework that generates MOF candidates in MOFid space, standardizes them, checks validity, predicts properties with a Transformer model, stores structured memory, and explores via multi-island search using iterative generate–validate–evaluate–remember cycles.",{"name":82,"@type":73,"acceptedAnswer":83},"How was LEMO Agent evaluated and validated experimentally?",{"text":84,"@type":76},"It was evaluated on CH4/N2 and CO2/N2 separation tasks, compared to generative/optimization/agent baselines, reconstructed and assessed using GCMC simulations, then down-selected by chemical feasibility and ligand purchasability for initial wet-lab synthesis and SEM characterization.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":22,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},"General","general"]