Characterizing Conical Intersections in DNA/RNA Nucleobases with Multiconfigurational Wave Functions of Varying...

Area of Science:

• Computational Chemistry
• Quantum Chemistry
• Photochemistry

Background:

• Conical intersections (CIs) are critical for excited-state dynamics in DNA/RNA nucleobases.
• Understanding CI topographies is essential for explaining photoinduced phenomena like DNA repair and photodamage.
• Previous studies have not systematically investigated the impact of computational model size on CI topography.

Purpose of the Study:

• To characterize photochemically relevant conical intersections in DNA/RNA nucleobases.
• To investigate the influence of active space size and basis set choice on CI topographies.
• To determine the key factors governing the accuracy of CI descriptions in molecular photochemistry.

Main Methods:

• Employed Cholesky decomposition-based complete active space self-consistent field (CASSCF) algorithms.
• Benchmarked various active space sizes and basis set contractions for different nucleobases and CI types.
• Analyzed the resulting CI topographies and their sensitivity to electronic correlation.

Main Results:

• CI topographies are highly sensitive to the amount and type of electron correlation included in the model.
• Changes in active space size did not show a converging pattern for CI topographies.
• Analogous topographies were observed for CIs mediating population transfer to dark states, but not for the 'ethylene-like' CI involved in ultrafast decay.

Conclusions:

• Accurate description of electronic states at CIs is more critical than structural factors for classification.
• Careful selection of the active space is essential for accurately modeling CI topographies in photochemistry.
• The findings highlight the importance of electron correlation in describing excited-state dynamics of nucleobases.