Pub Date : 2026-03-01Epub Date: 2026-02-05DOI: 10.1016/j.coche.2026.101228
Angan Mukherjee, Victor M Zavala
Physics-constrained machine learning (PCML) combines physical models with data-driven approaches to improve reliability, generalizability, and interpretability. Although PCML has shown significant benefits in diverse scientific and engineering domains, technical and intellectual challenges hinder its applicability in complex chemical engineering applications. Key difficulties include determining the amount and type of physical knowledge to embed, designing effective fusion strategies with machine learning, scaling models to large datasets and simulators, and quantifying predictive uncertainty. This perspective summarizes recent developments and highlights challenges/opportunities in applying PCML to chemical engineering, emphasizing closed-loop experimental design, real-time dynamics and control, and handling of multiscale phenomena.
{"title":"Physics-constrained machine learning for chemical engineering","authors":"Angan Mukherjee, Victor M Zavala","doi":"10.1016/j.coche.2026.101228","DOIUrl":"10.1016/j.coche.2026.101228","url":null,"abstract":"<div><div>Physics-constrained machine learning (PCML) combines physical models with data-driven approaches to improve reliability, generalizability, and interpretability. Although PCML has shown significant benefits in diverse scientific and engineering domains, technical and intellectual challenges hinder its applicability in complex chemical engineering applications. Key difficulties include determining the amount and type of physical knowledge to embed, designing effective fusion strategies with machine learning, scaling models to large datasets and simulators, and quantifying predictive uncertainty. This perspective summarizes recent developments and highlights challenges/opportunities in applying PCML to chemical engineering, emphasizing closed-loop experimental design, real-time dynamics and control, and handling of multiscale phenomena.</div></div>","PeriodicalId":292,"journal":{"name":"Current Opinion in Chemical Engineering","volume":"51 ","pages":"Article 101228"},"PeriodicalIF":6.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147394684","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2025-12-24DOI: 10.1016/j.coche.2025.101209
Sophia Rupprecht, Qinghe Gao, Tanuj Karia, Artur M Schweidtmann
Large language model (LLM)-based multi-agent systems (MASs) are a recent but rapidly evolving technology with the potential to transform chemical engineering by decomposing complex workflows into teams of collaborative agents with specialized knowledge and tools. This review surveys the state-of-the-art of MASs within chemical engineering. While early studies demonstrate promising results, scientific challenges remain, including the design of tailored architectures, integration of heterogeneous data modalities, development of foundation models with domain-specific modalities, and strategies for ensuring transparency, safety, and environmental impact. As a young but fast-moving field, MASs offer exciting opportunities to rethink chemical engineering workflows.
{"title":"Multi-agent systems for chemical engineering: a review and perspective","authors":"Sophia Rupprecht, Qinghe Gao, Tanuj Karia, Artur M Schweidtmann","doi":"10.1016/j.coche.2025.101209","DOIUrl":"10.1016/j.coche.2025.101209","url":null,"abstract":"<div><div>Large language model (LLM)-based multi-agent systems (MASs) are a recent but rapidly evolving technology with the potential to transform chemical engineering by decomposing complex workflows into teams of collaborative agents with specialized knowledge and tools. This review surveys the state-of-the-art of MASs within chemical engineering. While early studies demonstrate promising results, scientific challenges remain, including the design of tailored architectures, integration of heterogeneous data modalities, development of foundation models with domain-specific modalities, and strategies for ensuring transparency, safety, and environmental impact. As a young but fast-moving field, MASs offer exciting opportunities to rethink chemical engineering workflows.</div></div>","PeriodicalId":292,"journal":{"name":"Current Opinion in Chemical Engineering","volume":"51 ","pages":"Article 101209"},"PeriodicalIF":6.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145836473","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The strategies to increase the hydrogen production capabilities of the current state-of-the-art systems, such as the anion exchange membrane water electrolyzer (AEMWE), are gaining considerable attention. Despite steady progress being made with the green hydrogen-producing capability, the AEMWE technology is facing durability issues over extended operations at the stack level. This review is primarily focused on the degradation and durability challenges in AEM. The degradation modes in AEM include: i) chemical and mechanical degradation of the polymer backbones during dry and wet operations. ii) Degradation of quaternary ammonium headgroups in the AEM due to hydroxyl radical attack. iii) Membrane swelling due to increased water uptake within the membrane electrode assembly. From an industrial viewpoint, this review discusses the latest developments on durable AEM design, structure–property relationships, systematic monitoring of the degradation pathways and key mitigation strategies. The critical viewpoints highlighted in this review would advance the fundamental understanding and engineering of next-generation AEMs for deployment in AEMWE at the industrial level.
{"title":"Addressing degradation and durability challenges in anion exchange membranes for advancing anion exchange membrane water electrolyzers","authors":"Arun Prakash Periasamy , N Clament Sagaya Selvam , Balamurugan Devadas","doi":"10.1016/j.coche.2026.101226","DOIUrl":"10.1016/j.coche.2026.101226","url":null,"abstract":"<div><div>The strategies to increase the hydrogen production capabilities of the current state-of-the-art systems, such as the anion exchange membrane water electrolyzer (AEMWE), are gaining considerable attention. Despite steady progress being made with the green hydrogen-producing capability, the AEMWE technology is facing durability issues over extended operations at the stack level. This review is primarily focused on the degradation and durability challenges in AEM. The degradation modes in AEM include: i) chemical and mechanical degradation of the polymer backbones during dry and wet operations. ii) Degradation of quaternary ammonium headgroups in the AEM due to hydroxyl radical attack. iii) Membrane swelling due to increased water uptake within the membrane electrode assembly. From an industrial viewpoint, this review discusses the latest developments on durable AEM design, structure–property relationships, systematic monitoring of the degradation pathways and key mitigation strategies. The critical viewpoints highlighted in this review would advance the fundamental understanding and engineering of next-generation AEMs for deployment in AEMWE at the industrial level.</div></div>","PeriodicalId":292,"journal":{"name":"Current Opinion in Chemical Engineering","volume":"51 ","pages":"Article 101226"},"PeriodicalIF":6.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146073406","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2025-12-12DOI: 10.1016/j.coche.2025.101204
Jesús Serrano-Jiménez, Carlos Martín, Marina Pinzón, Paula Sánchez, Ana Raquel de la Osa
Hydrogen production via ammonia electrolysis is a promising alternative but still faces significant challenges hindering its practical deployment. Research has primarily focused on the development of anodic electrocatalysts, while limited studies address their integration into a functional laboratory-scale electrolyzer. Indeed, only two notable large-scale investigations have been reported. This review provides insight into the advances achieved in the past five years in the field of ammonia electrolysis, encompassing both small- and large-scale systems. First, the ongoing trends in the design of noble and non-noble electrocatalysts to maximize activity and stability have been analyzed. At the laboratory scale, the influence of crucial elements (porous transport layers and membranes) as well as the operational parameters (such as feed composition, temperature, etc.) is discussed, highlighting the most promising approaches to improve overall electrolyzer performance. At a larger scale, the main achievements are associated with system scale-up, particularly those related to stack design and material availability.
{"title":"Exploring the potential of ammonia electrolysis for hydrogen production: from lab-performance to stack architectures","authors":"Jesús Serrano-Jiménez, Carlos Martín, Marina Pinzón, Paula Sánchez, Ana Raquel de la Osa","doi":"10.1016/j.coche.2025.101204","DOIUrl":"10.1016/j.coche.2025.101204","url":null,"abstract":"<div><div>Hydrogen production via ammonia electrolysis is a promising alternative but still faces significant challenges hindering its practical deployment. Research has primarily focused on the development of anodic electrocatalysts, while limited studies address their integration into a functional laboratory-scale electrolyzer. Indeed, only two notable large-scale investigations have been reported. This review provides insight into the advances achieved in the past five years in the field of ammonia electrolysis, encompassing both small- and large-scale systems. First, the ongoing trends in the design of noble and non-noble electrocatalysts to maximize activity and stability have been analyzed. At the laboratory scale, the influence of crucial elements (porous transport layers and membranes) as well as the operational parameters (such as feed composition, temperature, etc.) is discussed, highlighting the most promising approaches to improve overall electrolyzer performance. At a larger scale, the main achievements are associated with system scale-up, particularly those related to stack design and material availability.</div></div>","PeriodicalId":292,"journal":{"name":"Current Opinion in Chemical Engineering","volume":"51 ","pages":"Article 101204"},"PeriodicalIF":6.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145733691","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2025-12-19DOI: 10.1016/j.coche.2025.101207
Zhifei Yuliu, Ruofan Shi, Marianthi G Ierapetritou
Optimizing supply chains is essential for the efficient and successful operation of chemical engineering processes. Real-world supply chain performance is frequently impacted by uncertainty. This work reviews recent literature in supply chain design and planning under uncertainty, considering two or multiple stages. Progress in various applications such as biomass valorization, waste management, and pharmaceutical manufacturing is examined. Recently used uncertainty handling methods (scenario-based stochastic programming, distributionally robust optimization, fuzzy programming, and chance-constrained programming) and computational challenges are discussed. Key gaps identified for future research include the treatment of decision-dependent uncertainty and solution strategies for nonconvexity.
{"title":"Mathematical programming approaches to supply chain optimization under uncertainty: a review","authors":"Zhifei Yuliu, Ruofan Shi, Marianthi G Ierapetritou","doi":"10.1016/j.coche.2025.101207","DOIUrl":"10.1016/j.coche.2025.101207","url":null,"abstract":"<div><div>Optimizing supply chains is essential for the efficient and successful operation of chemical engineering processes. Real-world supply chain performance is frequently impacted by uncertainty. This work reviews recent literature in supply chain design and planning under uncertainty, considering two or multiple stages. Progress in various applications such as biomass valorization, waste management, and pharmaceutical manufacturing is examined. Recently used uncertainty handling methods (scenario-based stochastic programming, distributionally robust optimization, fuzzy programming, and chance-constrained programming) and computational challenges are discussed. Key gaps identified for future research include the treatment of decision-dependent uncertainty and solution strategies for nonconvexity.</div></div>","PeriodicalId":292,"journal":{"name":"Current Opinion in Chemical Engineering","volume":"51 ","pages":"Article 101207"},"PeriodicalIF":6.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145787467","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-01-14DOI: 10.1016/j.coche.2025.101222
Fabian Hauf , Sam Van Daele , Mulatu Kassie Birhanu , Stefan Haufe , Tom Breugelmans , Elias Klemm
The electrochemical reduction reaction of CO2 into valuable chemicals offers a promising route for carbon management and renewable energy storage. However, the economic feasibility of conventional CO2 electrolysis is hindered by the energy-intensive provision of CO2. This mini review systematically explores the integrated CO2 electrolysis approach, which directly couples the electrolyzer with CO2 capture media, thereby passing costly intermediate stages. We concisely review the core aspects of this technology, including specialized cell designs, critical process parameters, and a focused comparison of absorbent solutions. This comparison encompasses their CO2 absorption capacity, as well as the advantages and limitations of each group of absorbents in the integrated system. The role of reactive capture of amino acids in integrated electrolysis is also highlighted briefly. Finally, the review aims to assess the technology readiness level of the integrated CO2 electrolysis and guide future research towards the development of an efficient and scalable technology.
{"title":"Integrated CO2 capture and electrolysis: advancing industrial implementation","authors":"Fabian Hauf , Sam Van Daele , Mulatu Kassie Birhanu , Stefan Haufe , Tom Breugelmans , Elias Klemm","doi":"10.1016/j.coche.2025.101222","DOIUrl":"10.1016/j.coche.2025.101222","url":null,"abstract":"<div><div>The electrochemical reduction reaction of CO<sub>2</sub> into valuable chemicals offers a promising route for carbon management and renewable energy storage. However, the economic feasibility of conventional CO<sub>2</sub> electrolysis is hindered by the energy-intensive provision of CO<sub>2</sub>. This mini review systematically explores the integrated CO<sub>2</sub> electrolysis approach, which directly couples the electrolyzer with CO<sub>2</sub> capture media, thereby passing costly intermediate stages. We concisely review the core aspects of this technology, including specialized cell designs, critical process parameters, and a focused comparison of absorbent solutions. This comparison encompasses their CO<sub>2</sub> absorption capacity, as well as the advantages and limitations of each group of absorbents in the integrated system. The role of reactive capture of amino acids in integrated electrolysis is also highlighted briefly. Finally, the review aims to assess the technology readiness level of the integrated CO<sub>2</sub> electrolysis and guide future research towards the development of an efficient and scalable technology.</div></div>","PeriodicalId":292,"journal":{"name":"Current Opinion in Chemical Engineering","volume":"51 ","pages":"Article 101222"},"PeriodicalIF":6.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145972960","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-02-17DOI: 10.1016/j.coche.2026.101232
Andrew J Medford , Todd N Whittaker , Bjarne Kreitz , David W Flaherty , John R Kitchin
Heterogeneous catalysis research struggles to connect intrinsic kinetics with experimentally observed behavior due to complex multiscale models, limited observability, and a many-to-one mapping between mechanisms and data. Advances in operando experiments, atomic-scale models, microkinetic models, and reactor simulations provide rich information, but dramatically expand model complexity and uncertainty. Artificial intelligence can reduce the human time needed for modeling by enabling ‘self-driving’ multiscale models that automate model construction, refinement, and validation across scales. Increased throughput will result in large ensembles of multiscale models that better explore parameter space, yield insight into sensitivity and uncertainty, and improve quantitative agreement between theory and experiment.
{"title":"Prospects for using artificial intelligence to understand intrinsic kinetics of heterogeneous catalytic reactions","authors":"Andrew J Medford , Todd N Whittaker , Bjarne Kreitz , David W Flaherty , John R Kitchin","doi":"10.1016/j.coche.2026.101232","DOIUrl":"10.1016/j.coche.2026.101232","url":null,"abstract":"<div><div>Heterogeneous catalysis research struggles to connect intrinsic kinetics with experimentally observed behavior due to complex multiscale models, limited observability, and a many-to-one mapping between mechanisms and data. Advances in <em>operando</em> experiments, atomic-scale models, microkinetic models, and reactor simulations provide rich information, but dramatically expand model complexity and uncertainty. Artificial intelligence can reduce the human time needed for modeling by enabling ‘self-driving’ multiscale models that automate model construction, refinement, and validation across scales. Increased throughput will result in large ensembles of multiscale models that better explore parameter space, yield insight into sensitivity and uncertainty, and improve quantitative agreement between theory and experiment.</div></div>","PeriodicalId":292,"journal":{"name":"Current Opinion in Chemical Engineering","volume":"51 ","pages":"Article 101232"},"PeriodicalIF":6.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147394686","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-02-25DOI: 10.1016/j.coche.2026.101236
Hans Hasse, Sebastian Schmitt, Fabian Jirasek
Artificial intelligence is currently transforming thermodynamics. Hybrid models that combine machine learning (ML) with physical modeling enable predictions of thermophysical properties with unprecedented scope and accuracy. Focusing on the thermophysical properties of fluids, recent advances in this field are highlighted, covering two hybridization techniques: (i) embedding ML into physical models and (ii) incorporating physical knowledge into ML models. The discussion covers different types of thermodynamic models: (i) excess Gibbs energy models, (ii) equations of state, and (iii) force field models. The new hybrid models combine the soundness of physical models with the flexibility of ML models and give the best results when trained on large data sets, which are, however, not always available. The new hybrid models often significantly outperform widely used classical physical thermodynamic benchmark models. We have only begun to explore the new routes opened up by hybrid thermodynamic modeling; this review provides a starting point for future work in this field.
{"title":"Artificial intelligence in thermodynamics: hybrid modeling of thermophysical properties of fluids","authors":"Hans Hasse, Sebastian Schmitt, Fabian Jirasek","doi":"10.1016/j.coche.2026.101236","DOIUrl":"10.1016/j.coche.2026.101236","url":null,"abstract":"<div><div>Artificial intelligence is currently transforming thermodynamics. Hybrid models that combine machine learning (ML) with physical modeling enable predictions of thermophysical properties with unprecedented scope and accuracy. Focusing on the thermophysical properties of fluids, recent advances in this field are highlighted, covering two hybridization techniques: (i) embedding ML into physical models and (ii) incorporating physical knowledge into ML models. The discussion covers different types of thermodynamic models: (i) excess Gibbs energy models, (ii) equations of state, and (iii) force field models. The new hybrid models combine the soundness of physical models with the flexibility of ML models and give the best results when trained on large data sets, which are, however, not always available. The new hybrid models often significantly outperform widely used classical physical thermodynamic benchmark models. We have only begun to explore the new routes opened up by hybrid thermodynamic modeling; this review provides a starting point for future work in this field.</div></div>","PeriodicalId":292,"journal":{"name":"Current Opinion in Chemical Engineering","volume":"51 ","pages":"Article 101236"},"PeriodicalIF":6.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147394688","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2025-12-13DOI: 10.1016/j.coche.2025.101206
Rafael Granados-Fernández , Miguel A. Rodriguez-Cano , Cristina Sáez, Justo Lobato, Manuel A. Rodrigo
Scaling up electrochemical technologies is key to industrial adoption. Mature processes like chlor-alkali and alumina electrolysis continue evolving to address environmental concerns. Recent commercial advances include water electrolysis and hydrogen fuel cells, though efficiency improvements are needed for profitability. Most emerging technologies face challenges in stability, reproducibility, and operability under real-world conditions. Environmental treatments remain limited to demonstration scale, while CO₂ electroreduction and organic electrosynthesis are still in early development. However, growing pressure to reduce emissions and replace fossil fuels is accelerating interest in e-fuels and electro-organic production. This review highlights recent progress and key barriers in scaling electrochemical technologies toward full-scale industrial implementation.
{"title":"Recent progress in scaling up promising electrochemical technologies","authors":"Rafael Granados-Fernández , Miguel A. Rodriguez-Cano , Cristina Sáez, Justo Lobato, Manuel A. Rodrigo","doi":"10.1016/j.coche.2025.101206","DOIUrl":"10.1016/j.coche.2025.101206","url":null,"abstract":"<div><div>Scaling up electrochemical technologies is key to industrial adoption. Mature processes like chlor-alkali and alumina electrolysis continue evolving to address environmental concerns. Recent commercial advances include water electrolysis and hydrogen fuel cells, though efficiency improvements are needed for profitability. Most emerging technologies face challenges in stability, reproducibility, and operability under real-world conditions. Environmental treatments remain limited to demonstration scale, while CO₂ electroreduction and organic electrosynthesis are still in early development. However, growing pressure to reduce emissions and replace fossil fuels is accelerating interest in e-fuels and electro-organic production. This review highlights recent progress and key barriers in scaling electrochemical technologies toward full-scale industrial implementation.</div></div>","PeriodicalId":292,"journal":{"name":"Current Opinion in Chemical Engineering","volume":"51 ","pages":"Article 101206"},"PeriodicalIF":6.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145733692","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2025-12-23DOI: 10.1016/j.coche.2025.101217
Sven M Papidocha , Andreas Burger , Varinia Bernales , Alán Aspuru-Guzik
The design of small molecules with tailored properties is a central goal in chemistry and materials science. Recent advances in machine learning provide powerful tools to accelerate the pace of discovery. One promising avenue for acceleration involves the use of generative models that propose novel candidates for diverse optimization tasks. Despite their promise, these methods are often evaluated solely using computational benchmarks, and many studies fail to advance proposed candidates to experimental validation in the wet lab. A key reason for this gap, the elephant in the room, is the limited synthesizability of the generated molecules. In response, the community has recently developed various strategies to address this challenge and incorporate synthesizability into generative design workflows. In this opinion, we provide a comprehensive overview of recent contributions that explicitly tackle molecular synthesizability, highlighting notable advances. We also discuss key limitations of current approaches and outline promising directions for future research.
{"title":"The elephant in the lab: synthesizability in generative small-molecule design","authors":"Sven M Papidocha , Andreas Burger , Varinia Bernales , Alán Aspuru-Guzik","doi":"10.1016/j.coche.2025.101217","DOIUrl":"10.1016/j.coche.2025.101217","url":null,"abstract":"<div><div>The design of small molecules with tailored properties is a central goal in chemistry and materials science. Recent advances in machine learning provide powerful tools to accelerate the pace of discovery. One promising avenue for acceleration involves the use of generative models that propose novel candidates for diverse optimization tasks. Despite their promise, these methods are often evaluated solely using computational benchmarks, and many studies fail to advance proposed candidates to experimental validation in the wet lab. A key reason for this gap, the elephant in the room, is the limited <em>synthesizability</em> of the generated molecules. In response, the community has recently developed various strategies to address this challenge and incorporate synthesizability into generative design workflows. In this opinion, we provide a comprehensive overview of recent contributions that explicitly tackle molecular synthesizability, highlighting notable advances. We also discuss key limitations of current approaches and outline promising directions for future research.</div></div>","PeriodicalId":292,"journal":{"name":"Current Opinion in Chemical Engineering","volume":"51 ","pages":"Article 101217"},"PeriodicalIF":6.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145836472","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号