Periodic Reporting for period 1 - INTDKIN (Reaction Mechanism of Methanol Conversion in Zeolite by Integrated Diffusion/Reaction Kinetics Model)
Reporting period: 2024-01-01 to 2026-03-31
Summary of the context and overall objectives of the project
The INTDKIN project was developed in response to several long-standing scientific and technological challenges in heterogeneous catalysis, particularly in the field of zeolite-catalyzed C1 chemistry and sustainable carbon conversion. Zeolite catalysts are widely used in industrial processes such as methanol-to-hydrocarbons (MTH), syngas conversion, and CO2 reduction because of their strong acidity, shape selectivity, and confined microporous structures. These catalytic technologies are highly relevant for the transition toward sustainable chemical production and low-carbon energy systems, especially within the broader context of carbon-neutral and circular-carbon strategies promoted by the European Green Deal and related international sustainability initiatives. Despite decades of research, the fundamental reaction mechanisms governing methanol and CO2 conversion in zeolites remain incompletely understood. One of the major unresolved problems is the difficulty in quantitatively describing the strong coupling between molecular diffusion and chemical reactions inside zeolite micropores. Conventional static density functional theory (DFT) approaches have been highly successful in identifying elementary reaction pathways under idealized conditions, but they often neglect the dynamic nature of catalytic processes under realistic reaction environments. In particular, the diffusion and confinement effects of reactive oxygenate intermediates within zeolite channels can substantially influence reaction thermodynamics, kinetics, and catalytic selectivity, yet these effects are difficult to capture using traditional computational methods. This limitation has contributed to persistent discrepancies between theoretical predictions and experimental observations in many zeolite-catalyzed reactions.
The INTDKIN project aimed to address these challenges by developing an integrated diffusion/reaction kinetics model capable of describing diffusion–reaction entanglement and reaction mechanism during methanol conversion in zeolites at the quantum-mechanical level. The overall objective of the project was to establish a new computational methodology that combines ab initio molecular dynamics (AIMD), machine-learning-potential molecular dynamics (MLP-MD), enhanced sampling techniques, and free-energy-based kinetic analysis to investigate catalytic reaction mechanisms under realistic reaction conditions. By moving beyond conventional static calculations toward dynamic simulations, the project sought to provide a more realistic and comprehensive description of catalytic processes occurring in confined zeolite environments. A second major objective of the project was to apply the developed methodology to important reactions in sustainable C1 chemistry, including methanol-to-hydrocarbons conversion, syngas conversion, and CO2-to-hydrocarbons processes. Through these studies, the project aimed to uncover the dynamic evolution of key oxygenate intermediates within zeolite micropores and clarify how diffusion–reaction coupling governs catalytic selectivity and reactivity. In particular, the project addressed fundamental scientific questions related to the induction period of methanol-to-hydrocarbons conversion and proposed a formaldehyde-mediated carbon–carbon bond formation mechanism, contributing new mechanistic understanding to one of the long-standing debates in zeolite catalysis.
The project pathway to impact was built around the development and dissemination of advanced computational methodologies capable of bridging the gap between theoretical simulations and experimentally observed catalytic behavior. By enabling quantitative descriptions of diffusion–reaction entanglement, the project provides researchers with new tools for understanding catalytic processes under realistic operating conditions. The developed framework is expected to support the rational design, optimization, and screening of zeolite catalysts with improved catalytic efficiency, product selectivity, and energy utilization. The methodology is broadly applicable not only to methanol and CO2 conversion, but also to wider classes of heterogeneous catalytic systems involving confined reactions and dynamic catalytic environments.
The expected impacts of the project extend across scientific, technological, and strategic dimensions. Scientifically, the project contributes to advancing the state of the art in computational heterogeneous catalysis by integrating dynamic simulations, enhanced sampling methods, machine-learning-assisted modeling, and high-performance computing into a unified framework. Technologically, the project supports the development of cleaner and more efficient catalytic processes relevant to sustainable fuel and chemical production. From a broader strategic perspective, the project aligns strongly with European priorities related to sustainable carbon utilization, digital transformation, and advanced computational technologies. In particular, the integration of machine learning and large-scale simulations contributes to the development of digital and data-driven approaches in materials and catalysis research, supporting the strategic objective of strengthening Europe’s leadership in sustainable and digital technologies. The project results are also relevant to a wide range of stakeholders, including academic researchers in catalysis and theoretical chemistry, industrial researchers in petrochemical and energy-related sectors, and developers of computational methodologies and digital tools for materials science. The dissemination of the project outcomes through high-impact scientific publications, international conferences, and research collaborations further strengthens the project’s pathway toward long-term scientific and technological impact.
The INTDKIN project aimed to address these challenges by developing an integrated diffusion/reaction kinetics model capable of describing diffusion–reaction entanglement and reaction mechanism during methanol conversion in zeolites at the quantum-mechanical level. The overall objective of the project was to establish a new computational methodology that combines ab initio molecular dynamics (AIMD), machine-learning-potential molecular dynamics (MLP-MD), enhanced sampling techniques, and free-energy-based kinetic analysis to investigate catalytic reaction mechanisms under realistic reaction conditions. By moving beyond conventional static calculations toward dynamic simulations, the project sought to provide a more realistic and comprehensive description of catalytic processes occurring in confined zeolite environments. A second major objective of the project was to apply the developed methodology to important reactions in sustainable C1 chemistry, including methanol-to-hydrocarbons conversion, syngas conversion, and CO2-to-hydrocarbons processes. Through these studies, the project aimed to uncover the dynamic evolution of key oxygenate intermediates within zeolite micropores and clarify how diffusion–reaction coupling governs catalytic selectivity and reactivity. In particular, the project addressed fundamental scientific questions related to the induction period of methanol-to-hydrocarbons conversion and proposed a formaldehyde-mediated carbon–carbon bond formation mechanism, contributing new mechanistic understanding to one of the long-standing debates in zeolite catalysis.
The project pathway to impact was built around the development and dissemination of advanced computational methodologies capable of bridging the gap between theoretical simulations and experimentally observed catalytic behavior. By enabling quantitative descriptions of diffusion–reaction entanglement, the project provides researchers with new tools for understanding catalytic processes under realistic operating conditions. The developed framework is expected to support the rational design, optimization, and screening of zeolite catalysts with improved catalytic efficiency, product selectivity, and energy utilization. The methodology is broadly applicable not only to methanol and CO2 conversion, but also to wider classes of heterogeneous catalytic systems involving confined reactions and dynamic catalytic environments.
The expected impacts of the project extend across scientific, technological, and strategic dimensions. Scientifically, the project contributes to advancing the state of the art in computational heterogeneous catalysis by integrating dynamic simulations, enhanced sampling methods, machine-learning-assisted modeling, and high-performance computing into a unified framework. Technologically, the project supports the development of cleaner and more efficient catalytic processes relevant to sustainable fuel and chemical production. From a broader strategic perspective, the project aligns strongly with European priorities related to sustainable carbon utilization, digital transformation, and advanced computational technologies. In particular, the integration of machine learning and large-scale simulations contributes to the development of digital and data-driven approaches in materials and catalysis research, supporting the strategic objective of strengthening Europe’s leadership in sustainable and digital technologies. The project results are also relevant to a wide range of stakeholders, including academic researchers in catalysis and theoretical chemistry, industrial researchers in petrochemical and energy-related sectors, and developers of computational methodologies and digital tools for materials science. The dissemination of the project outcomes through high-impact scientific publications, international conferences, and research collaborations further strengthens the project’s pathway toward long-term scientific and technological impact.
Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far
The INTDKIN project successfully developed an integrated diffusion/reaction kinetics framework to investigate methanol conversion in zeolites under realistic reaction conditions. By combining ab initio molecular dynamics (AIMD), machine-learning-potential molecular dynamics (MLP-MD), enhanced sampling techniques, and free-energy-based kinetic analysis, the project established a dynamic simulation methodology capable of describing the strong coupling between diffusion and catalytic reactions inside zeolite micropores. In WP1, the adsorption behavior of key oxygenate intermediates, including ketene, methyl ketene, and dimethyl ketene, was systematically investigated in SAPO-18 and MgAPO-18 zeolites. The work demonstrated that conventional static DFT calculations are insufficient to accurately describe the dynamic adsorption behavior of reactive intermediates under operando conditions, highlighting the importance of AIMD simulations and dynamic free-energy analyses. In WP2, the project focused on critical elementary reactions during methanol conversion. A new formaldehyde-mediated mechanism for the first carbon–carbon bond formation during methanol-to-hydrocarbons conversion was proposed and validated through both theoretical calculations and experimental observations. In addition, the dynamic evolution and protonation behavior of ketene intermediates in different zeolite frameworks were investigated using first-principles molecular dynamics simulations, revealing how framework composition and acidity influence olefin formation pathways. In WP3, the project investigated diffusion processes of ketenes and other oxygenates in zeolite frameworks. AIMD simulations revealed that ketene diffusion is strongly entangled with its reactivity toward Brønsted acid sites and guest molecules, demonstrating that diffusion and reaction cannot always be treated as independent processes. Further MLP-MD simulations showed that Brønsted acid sites can significantly facilitate the diffusion of many oxygenate intermediates inside SSZ-13 zeolites. Overall, the project successfully established an integrated reaction/diffusion kinetic model capable of simultaneously describing diffusion and reaction processes in zeolite catalysis. The project provided new mechanistic insights into methanol-to-hydrocarbons and CO2-to-hydrocarbons conversion and demonstrated that dynamic simulations offer substantial advantages over conventional static computational approaches for describing catalytic reactions in confined environments. The scientific outcomes of the project resulted in several high-impact publications and established a methodological foundation for future studies of heterogeneous catalysis under realistic reaction conditions.
Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)
The INTDKIN project delivered important advances beyond the current state of the art in both computational chemistry and zeolite catalysis by establishing an integrated diffusion/reaction kinetics framework for zeolite-catalyzed methanol conversion under realistic reaction conditions. Unlike conventional static DFT method, the developed methodology combines AIMD, MLP-MD, enhanced sampling techniques, and free-energy analysis to simultaneously describe diffusion and reaction processes in confined zeolite environments. A major scientific advancement of the project was the demonstration that diffusion and reaction of highly reactive intermediates, such as ketene, are strongly entangled and cannot always be treated as independent processes. This provides a new conceptual understanding of catalytic mechanisms in nanoporous materials. The project also proposed a new formaldehyde-mediated first carbon–carbon bond formation mechanism for methanol-to-hydrocarbons conversion, which was validated by both theoretical calculations and experimental observations. The developed methodologies provide powerful tools for the rational design and optimization of zeolite catalysts for sustainable C1 chemistry, including methanol and CO2 conversion. The integration of machine-learning-assisted molecular dynamics further improves the scalability and applicability of dynamic catalytic simulations. To ensure further uptake of the project results, future work should focus on extending the methodology toward larger-scale kinetic modeling, strengthening integration with operando experiments, and further developing machine-learning potentials for complex catalytic systems. The project results have been disseminated through high-impact publications and international collaborations, providing a strong foundation for future applications in sustainable catalysis and computational materials science.