[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-123523-en":3,"doc-seo-123523-105":30,"detail-sidebar-cat-0-en-105":92},{"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":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},123523,13056703020460,"Valentina","https://ap-avatar.wpscdn.com/avatar/be000253dac470eee5d?_k=1778207105932848923",8,"Research & Report","Machine learning potentials for redox chemistry in solution","Machine learning potentials (MLPs) are used to achieve quantum-mechanical accuracy with efficient atomistic simulations, but most local-energy models lack information about global system composition. This work introduces fourth-generation MLPs that correctly describe redox reactions in solution by recovering oxidation states and enabling electron-transfer behavior. Using ferrous and ferric ions in water, oxidation states are shown to match chloride counter-ion counts regardless of ion positions. The approach supports physically correct simulations of general redox chemistry in solution.","arXiv :2410 .03299v1 [physics .chem-ph] 4 Oct 2024  \nMachine learning potentials for redox chemistry in solution  \nEmir Kocer, 1, 2 Redouan El Haouari, 1, 2 Christoph Dellago,3 and J¨org Behler 1, 2, ∗  \n1 Lehrstuhl f¨ur Theoretische Chemie II, Ruhr-Universit¨at Bochum, 44780 Bochum, Germany  \n2 Research Center Chemical Sciences and Sustainability,  \nResearch Alliance Ruhr, 44780 Bochum, Germany  \n3 University of Vienna, Faculty of Physics, Boltzmanngasse 5, A-1090 Vienna, Austria  \nMachine learning potentials (MLPs) represent atomic interactions with quantum mechanical accuracy offering an efficient tool for atomistic simulations in many fields of science. However, most MLPs rely on local atomic energies without information about the global composition of the system. To date, this has prevented the application of MLPs to redox reactions in solution, which involve chemical species in different oxidation states and electron transfer between them. Here, we show that fourth-generation MLPs overcome this limitation and can provide a physically correct description of redox chemical reactions. For the example of ferrous (Fe2+ ) and ferric (Fe3+ ) ions in water we show that the correct oxidation states are obtained matching the number of chloride counter ions irrespective of their positions in the system. Moreover, we demonstrate that our method can describe electron-transfer processes between ferrous and ferric ions, paving the way to simulations of general redox chemistry in solution.  \nI. INTRODUCTION  \nRedox reactions, in which electrons are transferred from one chemical species to another, play a fundamental role across many fields of chemistry. Important examples are photosynthesis and enzymatic reactions that drive the processes of life, electrochemical water splitting for green hydrogen production, energy storage and conversion in batteries and fuel cells, as well as the corrosion of materials. Moreover, redox chemistry is considered as pivotal for the electrification of the chemical industry through transforming processes from fossil fuels to more sustainable renewable energy sources – an essential shift for maintaining the standard of living in modern societies [1] .  \nDue to the importance of redox reactions, a substantial effort has been devoted to understanding their underlying mechanisms in detail, and in recent years computer simulations have increasingly contributed to these efforts [2, 3] . To date, most of these simulations rely on accurate but computationally demanding quantum mechanical electronic structure methods, such as densityfunctional theory (DFT), to compute the atomic interactions. As a result, the length and time scales accessible in these ab initio molecular dynamics (AIMD) simulations remain limited, restricting their application to relatively simple model systems. More efficient potentials such as classical force fields [4, 5], which have been successfully employed in large-scale simulations in other fields of chemistry, are usually unable to describe electron transfer processes and the change of atomic oxidation states during the simulation. While a few advanced reactive force fields can overcome this limitation [6], they usually fall short of reaching “chemical accuracy” and thus lack the predictive power of electronic structure methods. For  \n∗ [joerg.behler@ruhr-uni-bochum.de](joerg.behler@ruhr-uni-bochum.de)  \nthese reasons, theoretical studies of complex redox reactions in solution have remained a significant challenge and are essentially confined to the domain of quantum chemistry.  \nIn recent years, rapid advances in machine learning techniques have driven a paradigm change in the construction of atomistic potentials. Rather than relying on the cumbersome development of increasingly sophisticated yet inherently approximate physical models, modern machine learning potentials (MLP) [7–11] “learn” the high-dimensional potential energy surface (PES)– which encapsulates all information about atomic","cbCaiumOpipQvAAP","https://ap.wps.com/l/cbCaiumOpipQvAAP","pdf",4988165,1,25,"English","en",105,"# Introduction\n## Limits of conventional simulation methods\n## Promise of machine learning potentials\n## Why redox in solution is challenging for MLPs\n## Proposed approach and demonstration","[{\"question\":\"What key limitation prevents most machine learning potentials from modeling redox reactions in solution?\",\"answer\":\"Most MLPs rely primarily on local atomic energies and cannot represent the global composition needed to distinguish different oxidation states and electron transfer in a fluctuating solvent environment.\"},{\"question\":\"How do the proposed fourth-generation MLPs address oxidation-state prediction?\",\"answer\":\"They overcome the lack of global composition information, enabling a physically correct description where oxidation states are recovered consistently using the relevant counter-ion information.\"},{\"question\":\"What example system is used to validate the method, and what is the main result?\",\"answer\":\"Ferrous (Fe2+) and ferric (Fe3+) ions in water. The method yields correct oxidation states that match chloride counter-ion counts independent of the ions’ positions.\"}]","Machine learning potentials for redox chemistry in solution | PDF",1785817090,63,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":28},"machine-learning-potentials-for-redox-chemistry-in-solution","",{"@graph":36,"@context":86},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"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":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/machine-learning-potentials-for-redox-chemistry-in-solution/123523/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05","2026-08-04",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What key limitation prevents most machine learning potentials from modeling redox reactions in solution?","Question",{"text":76,"@type":77},"Most MLPs rely primarily on local atomic energies and cannot represent the global composition needed to distinguish different oxidation states and electron transfer in a fluctuating solvent environment.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How do the proposed fourth-generation MLPs address oxidation-state prediction?",{"text":81,"@type":77},"They overcome the lack of global composition information, enabling a physically correct description where oxidation states are recovered consistently using the relevant counter-ion information.",{"name":83,"@type":74,"acceptedAnswer":84},"What example system is used to validate the method, and what is the main result?",{"text":85,"@type":77},"Ferrous (Fe2+) and ferric (Fe3+) ions in water. The method yields correct oxidation states that match chloride counter-ion counts independent of the ions’ positions.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]