What Controls the Quality of Photodynamical Simulations? Electronic Structure Versus Nonadiabatic Algorithm

Area of Science:

• Computational Chemistry
• Quantum Dynamics
• Photochemistry

Background:

• Nonadiabatic dynamics simulations are crucial for understanding photochemical processes.
• The choice of electronic structure method is known to influence simulation outcomes.
• Sensitivity analysis of electronic structure methods in nonadiabatic dynamics is often overlooked.

Purpose of the Study:

• To investigate the sensitivity of nonadiabatic dynamics to various electronic structure methods and algorithms.
• To analyze the nonadiabatic dynamics of cyclopropanone using a diverse set of computational approaches.
• To evaluate the impact of electronic structure choices on simulation accuracy.

Main Methods:

• Compared wave function-based (CASSCF, FOMO-CASCI, MS-CASPT2, XMS-CASPT2) and density-functional (REKS) methods.
• Included semiempirical (MRCI-OM3) electronic structure approaches.
• Utilized Landau-Zener surface hopping, fewest switches surface hopping, and ab initio multiple spawning algorithms.

Main Results:

• The electronic structure method choice significantly affects the accuracy of nonadiabatic dynamics for cyclopropanone.
• Potential energy surfaces showing qualitative and quantitative similarities did not guarantee consistent simulation results.
• No significant performance differences were observed among the tested nonadiabatic dynamics algorithms.

Conclusions:

• Relying solely on potential energy surface mapping for electronic structure selection can be misleading.
• Electronic structure method selection is critical for reliable computational photodynamics.
• The study highlights the need for rigorous electronic structure sensitivity analysis in nonadiabatic dynamics.