Deep Potential Molecular Dynamics Study of Propane Oxidative Dehydrogenation

Area of Science:

• Catalysis
• Chemical kinetics
• Computational chemistry

Background:

• Oxidative dehydrogenation (ODH) of light alkanes is vital for producing alkenes and oxygenated hydrocarbons.
• Controlling ODH mechanisms is crucial for maximizing target product selectivity (e.g., alkenes) and optimizing energy efficiency.
• Existing mechanistic understanding of propane ODH requires further elucidation for catalyst development.

Purpose of the Study:

• To elucidate the intricate reaction network and key mechanistic features of propane oxidative dehydrogenation (ODH).
• To investigate the influence of reaction pathways on propene selectivity and the role of hydrogen peroxide decomposition.
• To demonstrate the application and advantages of deep potential (DP) in studying complex reaction dynamics.

Main Methods:

• Employed deep potential (DP), a neural network atomic potential, for large-scale, accurate reactive dynamic simulations.
• Trained the DP model using a comprehensive dataset derived from density functional theory (DFT) calculations.
• Organized the elucidated reaction network hierarchically to highlight critical mechanistic steps.

Main Results:

• Revealed a more complex reaction mechanism for propane ODH compared to ethane ODH.
• Identified key features including propane and oxygen activation, propyl pathway influence on propene selectivity, and H2O2 decomposition.
• Demonstrated the efficacy of DP in accurately simulating complex reaction dynamics and mechanisms.

Conclusions:

• The study provides significant insights into the propane ODH mechanism, aiding in catalyst optimization.
• This work marks the first application of DP in ODH mechanistic studies, showcasing its power for complex systems.
• The findings are expected to guide the development of more efficient and selective ODH catalysts.