[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-126293-en":3,"doc-seo-126293-105":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":11,"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},126293,2336475104736,"Quinn","https://ap-avatar.wpscdn.com/avatar/22000c4c5e0e5b17e70?x-image-process=image/resize,m_fixed,w_180,h_180&k=1786591360781797222",8,"Research & Report","Modelling ligand exchange in metal complexes with machine learning potentials","Metal ions underpin structure and function across catalysis, charge transfer, and (bio)chemical assembly, yet predicting their structural and dynamic behavior in varied chemical environments remains difficult for both force-field and ab initio approaches. This article presents a computational strategy that trains machine learning potentials using MACE, an equivariant message-passing neural network, for metal–ligand complexes in explicit solvents. Applied to Mg2+ in water and Pd2+ in acetonitrile, the model reproduces equilibrium structures, coordination geometries, structural changes, and ligand-exchange free-energy barriers efficiently.","Open Access Article . Published on 03 August 2024. Downloaded on 2/28/2025 10:30:53 AM .  \nFaraday Discussions  \nCite this: Faraday Discuss., 2025, 256, 156  \n| PAPER | View Article Online View Journal | View Issue |\n| --- | --- |\n| Modelling ligand exchange in metal complexes with machine learning potentials†\u003Cbr>Veronika Juraskova,  ‡a Gers Tusha,  ‡b Hanwen Zhang,  a Lars V. Schfer  *b and Fernanda Duarte  *a\u003Cbr>Received 26th June 2024, Accepted 31st July 2024 DOI: 10.1039/d4fd00140k\u003Cbr>Metal ions are irreplaceable in many areas of chemistry, including (bio)catalysis, selfassembly and charge transfer processes. Yet, modelling their structural and dynamic properties in diverse chemical environments remains challenging for both force fieldsand ab initio methods. Here, we introduce a strategy to train machine learning potentials (MLPs) using MACE, an equivariant message-passing neural network, for metal–ligand complexes in explicit solvents. We explore the structure and ligand exchange dynamics of Mg2+ in water and Pd2+ in acetonitrile as two illustrative model systems. The trained potentials accurately reproduce equilibrium structures of the complexes in solution, including different coordination numbers and geometries. Furthermore, the MLPs can model structural changes between metal ions and ligands in the first coordination shell, and reproduce the free energy barriers for the corresponding ligand exchange. The strategy presented here provides a computationally efficient approach to model metal ions in solution, paving the way for modelling larger and more diverse metal complexes relevant to biomolecules and supramolecular assemblies.\u003Cbr>1 Introduction\u003Cbr>Metal ions have a central structural and functional role in many molecular systems, including catalysts, supramolecular assemblies, and biomolecules. Due to their relevance, much work has been done to investigate the structure, kinetics, and thermodynamic stability of metal complexes in solution, including the dynamics of metal–ligand exchange reactions.1 |  |\n\naChemistry Research Laboratory, University of Oxford, Oxford, OX1 3TA, UK. E-mail: fernanda. [duartegonzalez@chem.ox.ac.uk](duartegonzalez@chem.ox.ac.uk)  \nbCenter for Theoretical Chemistry, Ruhr University Bochum, D-44780 Bochum, Germany. E-mail: lars. [schaefer@ruhr-uni-bochum.de](schaefer@ruhr-uni-bochum.de)  \n† Electronic supplementary information (ESI) available. See DOI: [https://doi.org/10.1039/d4fd00140k](https://doi.org/10.1039/d4fd00140k)[ ](https://doi.org/10.1039/d4fd00140k)‡ Contributed equally to this work.  \n156 | Faraday Discuss., 2025, 256, 156–176 This journal is © The Royal Society of Chemistry 2025  \nOpen Access Article . Published on 03 August 2024. Downloaded on 2/28/2025 10:30:53 AM .  \nView Article Online  \nPaper Faraday Discussions  \nUsing a variety of experimental techniques, including X-ray absorption spectroscopy, neutron scattering and nuclear magnetic resonance (NMR) spectroscopy, several mechanisms have been proposed to describe ligand exchange in the 􀀁rst coordination shell of the metal ion. These mechanisms range from dissociative (D), involving an intermediate of lower coordination number, to associative (A), proceeding through an intermediate of higher coordination number. However, these are extreme cases – in most instances, no such idealised intermediate exists, and instead, a concerted interchange mechanism with dissociative (Id) or associate (Ia) characteristics occurs.2,3  \nOf particular interest is ligand exchange with solvent, with metal aqua complexes being the most extensively studied.4 The rate of this exchange depends on the nature of the metal ion, particularly ionic radii, charge, and coordination environment, ranging from 200 ps for Cs+ to 300 years for Ir3+ .4 Coordination with nonaqueous solvents such as alcohols, dimethyl sulfoxide (DMSO), acetonitrile (MeCN), and amides, has also been explored.5  \nAmong the cations investigated, signi􀀁cant efforts have been made to study","cbCaiglaS7omc9hJ","https://ap.wps.com/l/cbCaiglaS7omc9hJ","pdf",1260920,1,21,"English","en",105,"# Introduction\n## Mechanisms of ligand exchange in coordination shells\n## Solvent ligand exchange and relevant timescales\n## Case studies: Mg2+ in water and Pd2+ in acetonitrile","[{\"question\":\"What problem does the study address in modeling metal ions?\",\"answer\":\"It targets the challenge of accurately predicting metal-ion structural and dynamic properties in diverse chemical environments using either force fields or ab initio methods.\"},{\"question\":\"How is the machine learning potential constructed in the approach?\",\"answer\":\"It trains MLPs with MACE, an equivariant message-passing neural network, for metal–ligand complexes in explicit solvents.\"},{\"question\":\"Which systems are used to demonstrate the method and what does it reproduce?\",\"answer\":\"The study uses Mg2+ in water and Pd2+ in acetonitrile, reproducing equilibrium structures, coordination numbers and geometries, structural changes in the first coordination shell, and ligand-exchange free-energy barriers.\"}]","Modelling ligand exchange in metal complexes with machine learning potentials | PDF",1785904294,53,{"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},"modelling-ligand-exchange-in-metal-complexes-with-machine-learning-potentials","",{"@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/modelling-ligand-exchange-in-metal-complexes-with-machine-learning-potentials/126293/",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","2026-08-24","2026-08-05",true,{"@type":66,"interactionType":67,"userInteractionCount":11},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What problem does the study address in modeling metal ions?","Question",{"text":76,"@type":77},"It targets the challenge of accurately predicting metal-ion structural and dynamic properties in diverse chemical environments using either force fields or ab initio methods.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How is the machine learning potential constructed in the approach?",{"text":81,"@type":77},"It trains MLPs with MACE, an equivariant message-passing neural network, for metal–ligand complexes in explicit solvents.",{"name":83,"@type":74,"acceptedAnswer":84},"Which systems are used to demonstrate the method and what does it reproduce?",{"text":85,"@type":77},"The study uses Mg2+ in water and Pd2+ in acetonitrile, reproducing equilibrium structures, coordination numbers and geometries, structural changes in the first coordination shell, and ligand-exchange free-energy barriers.","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"]