Unlocking azobenzene isomerization mechanisms via an LLM agent-driven workflow integrating simulation, experiment, and machine learning

Key Points

• The research aims to clarify the mechanisms of Z↔E isomerization in azobenzene derivatives through an innovative workflow.
• Utilized a machine learning agent to design experiments and simulations.
• Conducted both spectroscopic analyses and computational modeling.
• Integrated findings from simulations with experimental data.
• Identified clear isomerization pathways for bridged azobenzene derivatives.
• Showed that the new workflow enhances understanding over traditional methods.
• Revealed key structural features that influence isomerization dynamics.