[{"data":1,"prerenderedAt":440},["ShallowReactive",2],{"blog-post-opendac-2025-the-largest-open-dataset-for-direct-air-capture":3},{"_id":4,"_updatedAt":5,"acknowledgements":6,"body":18,"credits":353,"excerpt":354,"hero":355,"publishedAt":370,"related":371,"seo":430,"slug":432,"thumbnail":48,"title":357,"headings":433,"footnotes":439,"hasMath":431},"post-medium-4f407afd45ea","2026-11-13T13:26:13Z",[7],{"_key":8,"_type":9,"children":10,"markDefs":16,"style":17},"3a9ed4d53cd3","block",[11],{"_key":12,"_type":13,"marks":14,"text":15},"3fa49389f30b","span",[],"With thanks to Meta FAIR, Georgia Tech, Carnegie Mellon University, Rice University, Oak Ridge National Laboratory and the whole CuspAI team who worked on this.",[],"normal",[19,51,70,78,88,96,108,120,128,137,145,153,161,180,189,197,222,234,246,258,267,275,283,291,299,308,322,336],{"_key":20,"_type":9,"children":21,"markDefs":45,"style":17},"7570c470e24b",[22,26,32,36,41],{"_key":23,"_type":13,"marks":24,"text":25},"12aee8f2100f",[],"Today, we’ve jointly published the largest open-access dataset to date for carbon capture materials in ",{"_key":27,"_type":13,"marks":28,"text":31},"f1edcec45ce6",[29,30],"strong","em","Nature Chemical Engineering.",{"_key":33,"_type":13,"marks":34,"text":35},"9ce3819b555e",[]," Working with Meta FAIR, Georgia Tech, and leading academic institutions*, we built the ",{"_key":37,"_type":13,"marks":38,"text":40},"faeadd91cfaa",[39],"f9589f761bf2","OpenDAC 2025",{"_key":42,"_type":13,"marks":43,"text":44},"7f73fd247c5a",[]," (ODAC25) dataset, which models how gases like CO2, N2, O2 and H2O interact with porous materials.",[46],{"_key":39,"_type":47,"internalSlug":48,"type":49,"url":50},"link",null,"external","https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs44286-026-00445-5",{"_key":52,"_type":9,"children":53,"markDefs":67,"style":17},"b003771a14a4",[54,58,63],{"_key":55,"_type":13,"marks":56,"text":57},"f999b354171b",[],"Direct Air Capture (DAC), pulling CO2 directly from the atmosphere, is an emerging technology with real promise for climate mitigation. It depends on materials including specialised porous materials called Metal-Organic Frameworks (MOFs), recognised with ",{"_key":59,"_type":13,"marks":60,"text":62},"cb88ea62a0a6",[61],"5cf6d3e732a1","the 2025 Nobel Prize in Chemistry",{"_key":64,"_type":13,"marks":65,"text":66},"035e71139426",[],", which act like atomic sponges.",[68],{"_key":61,"_type":47,"internalSlug":48,"type":49,"url":69},"https:\u002F\u002Fwww.nobelprize.org\u002Fprizes\u002Fchemistry\u002F2025\u002Fsummary\u002F",{"_key":71,"_type":9,"children":72,"markDefs":77,"style":17},"6b696a2f6d27",[73],{"_key":74,"_type":13,"marks":75,"text":76},"0d59d104c3a2",[],"While this study focuses on DAC, it also shows how generative AI and large-scale simulation can rapidly expand the frontiers of materials science across industries.",[],{"_key":79,"_type":9,"children":80,"markDefs":85,"style":86,"anchorId":87},"799a21bd58a8",[81],{"_key":82,"_type":13,"marks":83,"text":84},"9475fedf18e7",[],"Modelling materials under real-world complexity",[],"h2","modelling-materials-under-real-world-complexity",{"_key":89,"_type":9,"children":90,"markDefs":95,"style":17},"4f9bab3c1a7b",[91],{"_key":92,"_type":13,"marks":93,"text":94},"cd75b7fbb57e",[],"Traditionally, testing MOFs in simulations, or in real-world scenarios, has had major limitations:",[],{"_key":97,"_type":9,"children":98,"markDefs":107,"style":17},"d895b8b43438",[99,103],{"_key":100,"_type":13,"marks":101,"text":102},"9a52d1a9283e",[29],"Competition with humidity: ",{"_key":104,"_type":13,"marks":105,"text":106},"51087fcdd369",[],"Many MOFs look promising when exposed to CO2 alone, but struggle in real-world conditions because water vapour (H2O) in the air competes for the same binding sites. This can clog the MOF structures, degrade them permanently, or both.",[],{"_key":109,"_type":9,"children":110,"markDefs":119,"style":17},"d50cef4da6e2",[111,115],{"_key":112,"_type":13,"marks":113,"text":114},"97ee0b3bc28c",[29],"Computational bottlenecks:",{"_key":116,"_type":13,"marks":117,"text":118},"5a201f667b1b",[]," Simulating multi-gas interactions using traditional quantum physics methods — specifically Density Functional Theory (DFT), the gold-standard calculation for atomic behaviour — is extremely slow and computationally expensive.",[],{"_key":121,"_type":9,"children":122,"markDefs":127,"style":17},"5ccd74f8271b",[123],{"_key":124,"_type":13,"marks":125,"text":126},"2f22ecaabecd",[],"To accurately model how materials behave under real-world conditions, AI models require millions of high-precision quantum chemistry calculations. Previous datasets had notable gaps. They covered only CO2 and H2O, didn’t include chemically modified (functionalised) MOFs, and treated frameworks as more rigid than they really are.",[],{"_key":129,"_type":9,"children":130,"markDefs":135,"style":86,"anchorId":136},"f8b4ef79bbec",[131],{"_key":132,"_type":13,"marks":133,"text":134},"fe5ca88eb15c",[],"CuspAI’s role: expanding chemical space with generative AI",[],"cuspais-role-expanding-chemical-space-with-generative-ai",{"_key":138,"_type":9,"children":139,"markDefs":144,"style":17},"dd1f3b250cfd",[140],{"_key":141,"_type":13,"marks":142,"text":143},"15f74ad969c3",[],"For ODAC25, CuspAI contributed new, synthetic MOF structures, designed by one of its proprietary generative AI models.",[],{"_key":146,"_type":9,"children":147,"markDefs":152,"style":17},"a3983413b21d",[148],{"_key":149,"_type":13,"marks":150,"text":151},"7369f344623b",[],"Rather than limiting the search to materials that have already been made in the lab, the model generated 53,000 candidate frameworks. From these, we selected the 460 that most increased the dataset’s diversity, and computed them with DFT.",[],{"_key":154,"_type":9,"children":155,"markDefs":160,"style":17},"9619995d7d88",[156],{"_key":157,"_type":13,"marks":158,"text":159},"d6ee4da9923c",[],"These AI-designed MOFs extended the dataset toward more complex geometries, with larger pores, higher surface areas and lower densities than known experimental structures. This is evidence that generative models can usefully expand the range of structures a scientific dataset covers.",[],{"_key":162,"_type":9,"children":163,"markDefs":177,"style":17},"8fb32b8a7153",[164,168,173],{"_key":165,"_type":13,"marks":166,"text":167},"c91349b90b79",[],"CuspAI has also released an ",{"_key":169,"_type":13,"marks":170,"text":172},"abc3cd5ea51a",[171],"3c5f3aab4dfe","open-source software tool",{"_key":174,"_type":13,"marks":175,"text":176},"4041e84b0288",[]," that simulates how gases interact with materials. We used this tool to test our AI models against real-world data, to show that our AI can accurately predict how tightly a material will trap specific gases.",[178],{"_key":171,"_type":47,"internalSlug":48,"type":49,"url":179},"https:\u002F\u002Fzenodo.org\u002Frecords\u002F21537519",{"_key":181,"_type":9,"children":182,"markDefs":187,"style":86,"anchorId":188},"af9884e0c0f4",[183],{"_key":184,"_type":13,"marks":185,"text":186},"de4c829c1cad",[],"From quantum equations, to ML force fields",[],"from-quantum-equations-to-ml-force-fields",{"_key":190,"_type":9,"children":191,"markDefs":196,"style":17},"2845a846c536",[192],{"_key":193,"_type":13,"marks":194,"text":195},"79713ec444b1",[],"The ODAC25 dataset yielded:",[],{"_key":198,"_type":9,"children":199,"level":217,"listItem":218,"markDefs":219,"style":17},"38374442fe13",[200,204,208,213],{"_key":201,"_type":13,"marks":202,"text":203},"5ea41da24c7f",[29],"60 million DFT calculations: ",{"_key":205,"_type":13,"marks":206,"text":207},"0512ea2ff0af",[],"covering 15,000 MOFs, marking a substantial expansion over previous datasets (for example, ",{"_key":209,"_type":13,"marks":210,"text":212},"c535ed4ce376",[211],"e7c8ff029075","ODAC23",{"_key":214,"_type":13,"marks":215,"text":216},"44173f81b0fa",[]," included 38M calculations across 8,400 MOFs)",1,"bullet",[220],{"_key":211,"_type":47,"internalSlug":48,"type":49,"url":221},"https:\u002F\u002Fpubs.acs.org\u002Facscii\u002Farticle\u002F10\u002F5\u002F923\u002F144237\u002FThe-Open-DAC-2023-Dataset-and-Challenges-for",{"_key":223,"_type":9,"children":224,"level":217,"listItem":218,"markDefs":233,"style":17},"b73b84d21737",[225,229],{"_key":226,"_type":13,"marks":227,"text":228},"f684ae78f5d8",[29],"Multi-gas simulation: ",{"_key":230,"_type":13,"marks":231,"text":232},"3c4ac34d49af",[],"adsorption of CO2, H2O, N2 and O2, including CO2–H2O co-adsorption under humid conditions.",[],{"_key":235,"_type":9,"children":236,"level":217,"listItem":218,"markDefs":245,"style":17},"9de65a5954b7",[237,241],{"_key":238,"_type":13,"marks":239,"text":240},"3cd27523b07d",[29],"Accounting for MOF flexibility:",{"_key":242,"_type":13,"marks":243,"text":244},"fecffc4a6a05",[]," modeling structural deformation when gas molecules bind, which addresses a major limitation in simplified calculations that treat MOFs as rigid frameworks.",[],{"_key":247,"_type":9,"children":248,"level":217,"listItem":218,"markDefs":257,"style":17},"29fcdbd260b4",[249,253],{"_key":250,"_type":13,"marks":251,"text":252},"4dda4f700174",[29],"Training Machine Learning Force Fields ",{"_key":254,"_type":13,"marks":255,"text":256},"03f36d6bf160",[],"(MLFFs): by using this massive DFT dataset, researchers trained MLFFs, AI models that predict atomic forces and energies with quantum-level accuracy and at a fraction of the computational cost.",[],{"_key":259,"_type":9,"children":260,"markDefs":265,"style":86,"anchorId":266},"ab5d2faed8bc",[261],{"_key":262,"_type":13,"marks":263,"text":264},"37cb60e714af",[],"Proof that synthetic data scales",[],"proof-that-synthetic-data-scales",{"_key":268,"_type":9,"children":269,"markDefs":274,"style":17},"a34d5e22e398",[270],{"_key":271,"_type":13,"marks":272,"text":273},"f3c3c806ca5e",[],"The complete ODAC25 dataset, baseline models, and code are available open-source to empower researchers worldwide. These models enable scientists to quickly screen candidate materials, run atomistic simulations, and shortlist promising structures before ever stepping into a physical lab.",[],{"_key":276,"_type":9,"children":277,"markDefs":282,"style":17},"f702fb21e50d",[278],{"_key":279,"_type":13,"marks":280,"text":281},"5d6287a5eb6d",[],"Computational screening of adsorption is just the first step in real materials discovery, and additional simulations, synthesis, and testing is required to see which materials work in the real world.",[],{"_key":284,"_type":9,"children":285,"markDefs":290,"style":17},"b9821edad677",[286],{"_key":287,"_type":13,"marks":288,"text":289},"155662848d1c",[],"That said, ODAC25 demonstrates a scalable approach to materials design: expanding structural diversity beyond known experimental frameworks, followed by running high-accuracy simulations to map physical properties at scale.",[],{"_key":292,"_type":9,"children":293,"markDefs":298,"style":17},"ab4df84d9072",[294],{"_key":295,"_type":13,"marks":296,"text":297},"52534ed92640",[],"While applied here to porous materials, this approach provides a strong foundational template as we apply our generative engines to adjacent domains, such as advanced electronics and semiconductors.",[],{"_key":300,"_type":9,"children":301,"markDefs":306,"style":86,"anchorId":307},"75a3dfc34160",[302],{"_key":303,"_type":13,"marks":304,"text":305},"9706da4273c4",[],"Notes",[],"notes",{"_key":309,"_type":9,"children":310,"markDefs":320,"style":17},"a39c0d7ec7cc",[311,315],{"_key":312,"_type":13,"marks":313,"text":314},"118b1dd129f8",[29],"Read the paper: ",{"_key":316,"_type":13,"marks":317,"text":319},"69f9d5bfdd8a",[318],"322480cc4030","An open dataset of gas adsorption in deformable metal–organic frameworks for direct air capture",[321],{"_key":318,"_type":47,"internalSlug":48,"type":49,"url":50},{"_key":323,"_type":9,"children":324,"markDefs":334,"style":17},"f0ab0d38b208",[325,329],{"_key":326,"_type":13,"marks":327,"text":328},"103fe70714df",[29],"Access the dataset: ",{"_key":330,"_type":13,"marks":331,"text":333},"59666d735b4b",[332],"1ee463c68087","https:\u002F\u002Fhuggingface.co\u002Ffacebook\u002FODAC25",[335],{"_key":332,"_type":47,"internalSlug":48,"type":49,"url":333},{"_key":337,"_type":9,"children":338,"markDefs":351,"style":17},"b4f9a3c8d68c",[339,343,347],{"_key":340,"_type":13,"marks":341,"text":342},"daadca43b372",[29],"Code and supporting data:",{"_key":344,"_type":13,"marks":345,"text":346},"3387e1df731c",[]," ",{"_key":348,"_type":13,"marks":349,"text":179},"76efe39ad042",[350],"54d4227d5b19",[352],{"_key":350,"_type":47,"internalSlug":48,"type":49,"url":179},"Elias Moubarak, Pim de Haan, Sindy Löwe, Johann Brehmer","Today, we’ve jointly published the largest open-access dataset to date for carbon capture materials in Nature Chemical Engineering. Working with Meta FAIR, Georgia Tech, and leading academic…",{"caption":48,"credit":48,"gif":48,"image":356,"panelColor":368,"type":369,"video":48},{"alt":357,"asset":358,"crop":48,"hotspot":48},"OpenDAC 2025: the largest open dataset for direct air capture",{"_id":359,"metadata":360,"mimeType":366,"url":367},"image-63ee60b69350e431a64830e3bc72f421e0b25f98-2400x1348-jpg",{"dimensions":361,"lqip":365},{"aspectRatio":362,"height":363,"width":364},1.7804154302670623,1348,2400,"data:image\u002Fjpeg;base64,\u002F9j\u002F2wBDAAYEBQYFBAYGBQYHBwYIChAKCgkJChQODwwQFxQYGBcUFhYaHSUfGhsjHBYWICwgIyYnKSopGR8tMC0oMCUoKSj\u002F2wBDAQcHBwoIChMKChMoGhYaKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCj\u002FwAARCAALABQDASIAAhEBAxEB\u002F8QAGAAAAgMAAAAAAAAAAAAAAAAAAAMEBQf\u002FxAAhEAACAgEDBQEAAAAAAAAAAAABAgADERIhQQQFEzEycf\u002FEABYBAQEBAAAAAAAAAAAAAAAAAAMAAv\u002FEABkRAQEAAwEAAAAAAAAAAAAAAAEAERIhQf\u002FaAAwDAQACEQMRAD8Awbttmu5bWOWDb59kRnXUV1ZUEHnaVFLEAYPMa7t4mGo4gpnkSey3ZFbGISLb9D8hNaFX\u002F9k=","image\u002Fjpeg","https:\u002F\u002Fcdn.sanity.io\u002Fimages\u002Fho74vmlz\u002Fproduction\u002F63ee60b69350e431a64830e3bc72f421e0b25f98-2400x1348.jpg","teal","image","2026-10-29T16:06:17.000Z",[372,399,416],{"_id":373,"excerpt":374,"publishedAt":375,"slug":376,"thumbnail":377,"title":398},"25f54074-10d4-4f86-b855-6232676788e9","We're delighted to welcome Professor Susumu Kitagawa, winner of the 2025 Nobel Prize in Chemistry, as CuspAI’s latest advisor.","2026-11-19T10:00:00.000Z","nobel-laureate-susumu-kitagawa-joins-cuspais-advisory-board",{"alt":378,"asset":379,"crop":389,"hotspot":393},"Professor Susumu Kitagawa",{"_id":380,"metadata":381,"mimeType":387,"url":388},"image-d029c440c4e168db5800c4760636166bdfecbe12-3060x1722-png",{"dimensions":382,"lqip":386},{"aspectRatio":383,"height":384,"width":385},1.7770034843205575,1722,3060,"data:image\u002Fpng;base64,iVBORw0KGgoAAAANSUhEUgAAABQAAAALCAYAAAB\u002FCa1DAAAACXBIWXMAAAPoAAAD6AG1e1JrAAAC7klEQVQokVXS22+TdRzH8f47emcWjWHdwg41QYmpC8QLMHMLwQwSigGJikwEdkjECAluLDgXhoJZHTiVwE6Grufz0270fFiPtH3aPk\u002FbwZ2\u002Ft2nVqBeffO9e+eSTr6ZSkSmVSuTyWVLpGOFwEEly4vPZ8HqtuN0WXK4t\u002FAEnQZ8dk3EB4\u002FQ5Fic+Ym35LrlchrqqkMymkGIhoZHlGrJcpVwuUyjmSaXiPH0qEQi48fkdeL02XG4LTqcJv22T4G93sc5cwnjxJAvT53myuUYymyaeihKNh4WmWlOo1xVq9TrVapVCIUc8Huk03d7xEwh68PocWMzr2FcfEPz5Ns65KYwXxvjSMMLC\u002FE3sAQ9SRCIcCwpNrd5AUZsoavuqlMoldnfTJBJRotEQ4cg2wW0fDrsJ9\u002FoK0vK3WG9OcP\u002FSKWY\u002FPcH3i3P87jVjCznwRZxCU1caqI3W32lSkWVyuRzpdIpkMtGBI9EdpICbkGODyK+LOG9N8Wj6DPevfcba+hLWsBV7zIorahaadrt\u002FwRbVao1CoUA2myWTyZDZTZNs7xoJEvKZiT2+h3t+mpUJAz99M86WfQV\u002F2oEnYcUeMglNG2k092i2nndSr6s8K5UpFIrk8\u002FkOHE8k8AQktrY2cf1yB9PcZWbPvMfF0yPcvjeDRdrAHbFg2zELzX+x5t4L2hOUyjLF4jMSyTQ2h4eHj5\u002Fw44NVZm\u002Fd4frkZaY+PMbwW\u002F3oerQMjw7z1Y2rLD00smrd+D\u002FY2nuBojSoVKrk8kVMFjdXr3\u002FHJ+Nf8\u002FnkHIZzkwwdOspA3yCvv6qlq0vLvu4B3ta\u002Fy\u002FGTBsYnJv\u002Fa8B+0A6pN2r8ZT2X5YekRY4YvODJylrHTV3j\u002F+MfsH3yH1\u002Fbp6NMNMfDGIbQ9b9KtPUB3t46e3sH226hCbbREo7knmq3nQlGboiLXxHYoKWbml8Xo2AVx+MgpMfLBeXF09Kzo0w2J3n690B8+Jg7qh8X+fr3o6T0ourp6xcsvvfLHnw6JuYOIObvRAAAAAElFTkSuQmCC","image\u002Fpng","https:\u002F\u002Fcdn.sanity.io\u002Fimages\u002Fho74vmlz\u002Fproduction\u002Fd029c440c4e168db5800c4760636166bdfecbe12-3060x1722.png",{"_type":390,"bottom":391,"left":392,"right":391,"top":391},"sanity.imageCrop",0,0.38344882933238,{"_type":394,"height":217,"width":395,"x":396,"y":397},"sanity.imageHotspot",0.61655117066762,0.6917244146661901,0.5,"Nobel Laureate Susumu Kitagawa joins CuspAI's advisory board",{"_id":400,"excerpt":401,"publishedAt":402,"slug":403,"thumbnail":404,"title":415},"post-medium-f5603d8259dd","Today marks a milestone for the acceleration of materials discovery as we launch the ‘AI Materials Foundry’, a global network of data, labs, compute and scientific expertise for the design of new…","2026-07-20T06:04:59.000Z","launching-our-ai-materials-foundry-and-450-million-series-b",{"alt":405,"asset":406,"crop":48,"hotspot":48},"Launching our ‘AI Materials Foundry’ — and $450 million Series B",{"_id":407,"metadata":408,"mimeType":387,"url":414},"image-8c9dd3da26c30da6d4b83a57dbf5db2f5b76a715-2168x1080-png",{"dimensions":409,"lqip":413},{"aspectRatio":410,"height":411,"width":412},2.0074074074074075,1080,2168,"data:image\u002Fpng;base64,iVBORw0KGgoAAAANSUhEUgAAABQAAAAKCAYAAAC0VX7mAAAACXBIWXMAAAPoAAAD6AG1e1JrAAACvElEQVQokQXBfU8TBwDA4fsaS5aZuCyZcXMbiIDlxcooaaFUHVC0s7wEraNV7LFaWqHclTv6er3CcS0UoRZoAVeEbm6iIZroH2ZZ\u002FEy\u002FPY9wyeHhmz43X3QM8J3VztQ9L\u002FKcn3h4mshsAH9EwifnCagmStagqmucbRY52yuybSYJqybi0g7ibA7vyAzC1Ss2LN2D\u002FNRux3Ktj8C922gL06wtzZJWYoSVLLOZdeaN5+jGBgdaitOCxj\u002FPNCrFNKliBdV8gbJkEn4UQ7hw7luutVhxtNjot\u002FTweGqUouRjN+FnI7NIKpcnWdplpXpMubzLX4Us7zaznG5lqZdXOTh5yeHfpxweHVOr1hC+PHee1h9bcVzuwtXZzROfm634JDvxMYqJOXTDwKzW2Xt1xlG9zp\u002FrGV4XVBqmQmNL48ObQz5\u002FesXnjw3+e3+E0NzejNXahstuxet2IoemWJfuY0bvklPnyZUqGPsNysevqVb3qK0qvNAXqGWi7KUjHBgSjc04757L\u002FLufQBj71YX4yIsUnWZZCpJPzZNXIyQiD0knVPRnNZaKu4QSK0iKypaucLyR4o9ViXzEz8SQg9H+DmKTDioLYwi+8V94GnpAJv47emKONV3ByGsk4xKquoy6UiKo5PHMRPlNDFPIqJyUcrxcU0iH\u002FXRbWvjh4teMuzqRAkMIbmcboze68N2181T0oCyKyIvzhMQgD2eCBGNJxGWDGWWFOUWjpOc5KeqcFFLoUoje6120NX\u002FP+HAvAd8IgsfWxC3LRR78fAVpfIiof4IJtxOXrZNBp52x6SCP5SxPkmso2gb76xXelrc53TYxExJOex89nW14hgcYHr2FcOd6O97WJuSmq+j9wyxO3sHnbMZt+Yqb1kuMeDzcF6MEwjIxOUfd3OH9doW3myaFeIzBPhs93R3cvOGkt3+A\u002FwGo9ODJAa3VGwAAAABJRU5ErkJggg==","https:\u002F\u002Fcdn.sanity.io\u002Fimages\u002Fho74vmlz\u002Fproduction\u002F8c9dd3da26c30da6d4b83a57dbf5db2f5b76a715-2168x1080.png","Launching our ‘AI Materials Foundry’ — and $450 million Series B",{"_id":417,"excerpt":418,"publishedAt":419,"slug":420,"thumbnail":421,"title":429},"post-medium-42c8ec5afb11","Kemira and CuspAI have applied generative AI end-to-end to the design of new materials targeting the removal of PFAS from water at trace concentrations — becoming the first commercial partnership to…","2026-05-21T06:04:26.000Z","new-ai-designed-materials-show-promising-potential-to-remove-forever-chemicals-from-drinking",{"alt":422,"asset":423,"crop":48,"hotspot":48},"Kemira and CuspAI apply AI to the design of new materials for PFAS removal",{"_id":424,"metadata":425,"mimeType":387,"url":428},"image-adc9d557734e3a6c31e42bed60c0419142d14650-2168x1080-png",{"dimensions":426,"lqip":427},{"aspectRatio":410,"height":411,"width":412},"data:image\u002Fpng;base64,iVBORw0KGgoAAAANSUhEUgAAABQAAAAKCAYAAAC0VX7mAAAACXBIWXMAAAPoAAAD6AG1e1JrAAACw0lEQVQokT3M609bZQDA4fdtzzktPYd2HfTCpRdLKC0gji5YMqy94Ha6lbkApczBViYEkQhYNxIS3NSIQLJMthB2ydxIcDEWzRI2FxeXqGhi5j7oF\u002FWD\u002F8zPmBifP+ARf\u002F\u002F2nD9+2efZ0z3ubq9xopzmYLQOm8eJzefECHsI9aV5eazCa3OrmAur9E29Q9e5fsKlZuqyDppSXsLpAEaXjvhz\u002FwnP9z7nu3tXqF5b4v2F0+RyPTS3hnEG\u002FRxsayF69BS9ExV6z75FYmSQjuEkkaEAXlNHP6ziajdwx9xojRpif2eTn3c2+Pb6EvcvT7ExM8yFUp5sqoeGjjZ8iW7a8ic5NDxKPJ8lnI4SNv009DswEhaUVokasKLUK8gaibizPM2Tm5\u002Fw4901dtffZX26xHulAV4v5IlmUkTMLIk3xklPzdI9WKAxGcDX68R9yI7abEF6BBa3RDokQhOImWKOreVZftjZ4Nfd23xz6ypbH3\u002FExaVLjM4vUKzMM768wtCFy7w0cIy6eD2uqB09pGJxSkSNQNoEQhUIRSDMU0Vmzpb4bGWRZ9U7\u002FPX4K35\u002F9ID9h4\u002FZq37N7vYOW5\u002FeoDgxSVNnCw6\u002FHb1Jwea1YNEF8t\u002FQ\u002Fl8mBeLY\u002FBXMM1OUx0qsL85RvbbK0+3b\u002FLRb5fsvvuTB5iZrFyu8mkni9BnYDig4fAraAQvSLv8PpVUihEAUPrxPenaF1Pgc5sgEQ8VRymdO8\u002FZkmenz5xgZPM6RTBeBdg9OTw21Ph1XqBabS0NYBVKRWFUriqogpUQUlu+Rq1wn+eYHdJ6cpDUzQCzXTzzbR6S3i\u002FpOP654Ld72OoJtjQQ7Q\u002FjjQewuAyElVqsVwzBwu91omoZ45fwlkmOLdBwvEzxSIJg2aTmRJ2Jm8CZj2GMGjlYHkUSEw5keXkx10xB7Aa1WR1okqqri8\u002FkIhULous4\u002F2mpgggHqOAcAAAAASUVORK5CYII=","https:\u002F\u002Fcdn.sanity.io\u002Fimages\u002Fho74vmlz\u002Fproduction\u002Fadc9d557734e3a6c31e42bed60c0419142d14650-2168x1080.png","AI-Designed materials show potential promise to remove “forever chemicals” from drinking water",{"metaDescription":48,"metaTitle":48,"noIndex":431,"ogImageUrl":48},false,"opendac-2025-the-largest-open-dataset-for-direct-air-capture",[434,435,436,437,438],{"id":87,"text":84},{"id":136,"text":134},{"id":188,"text":186},{"id":266,"text":264},{"id":307,"text":305},[],1795092687760]