Prediction Challenge: Simulating Rydberg photoexcited cyclobutanone with surface hopping dynamics based on different...

Area of Science:

• Chemical Physics
• Computational Chemistry
• Quantum Dynamics

Background:

• Nonadiabatic dynamics govern molecular behavior after photoexcitation.
• Accurate prediction of these dynamics is crucial for understanding chemical reactions.
• Cyclobutanone serves as a model system for studying photochemical processes.

Purpose of the Study:

• To evaluate the predictive power of various computational chemistry methods for nonadiabatic dynamics.
• To simulate the nonadiabatic dynamics of cyclobutanone following excitation to the n → 3s Rydberg S2 state.
• To compare simulation results with experimental data to assess computational accuracy.

Main Methods:

• Decoherence-corrected fewest-switches surface hopping simulations.
• Employing diverse electronic structure methods: MCSCF, multireference semiempirical, TD-DFT, ADC, and coupled cluster.
• Utilizing a specialized MCSCF for dissociative channels and full/approximated nonadiabatic couplings.

Main Results:

• MCSCF dynamics predict slow S2 deactivation (~10 ps) and ultrafast S1-S0 transfer (<100 fs).
• CO elimination (C3 channel) is the dominant dissociation pathway.
• Other computational methods predict significantly shorter S2 lifetimes and C2H4 formation (C2 channel) predominance, contradicting MCSCF findings.

Conclusions:

• The specialized MCSCF method provides a more accurate description of cyclobutanone's nonadiabatic dynamics.
• Routine computational methods may lack sufficient predictive power for complex photochemical processes.
• Discrepancies highlight the need for advanced theoretical approaches in computational chemistry.