Development of Parallel On-the-Fly Algorithm for Global Exploration of Conical Intersection Seam Space.

Area of Science:

• Computational Chemistry
• Theoretical Chemistry
• Photochemistry

Background:

• Conical intersection (CI) seams are critical in molecular systems where electronic states degenerate, influencing photochemical processes.
• Nonradiative decay pathways often occur near minimum energy conical intersections (MECIs), necessitating their exploration.
• Existing CI seam exploration methods are often local, requiring good initial guesses and limiting global discovery.

Purpose of the Study:

• To develop a novel algorithm for the global exploration of conical intersection seam spaces.
• To overcome the computational expense and scalability limitations of traditional global search algorithms for large molecular systems.
• To enable automated discovery of minimum energy conical intersections (MECIs) crucial for understanding molecular photochemistry.

Main Methods:

• Development of a parallel on-the-fly algorithm for global CI seam exploration.
• Integration with on-the-fly energy evaluations using multireference electronic structure methods.
• Parallelization of the algorithm to enhance computational efficiency and scalability.

Main Results:

• The algorithm successfully identified all known MECIs and several new ones for photostatin (PST) and stilbene.
• Application to butadiene revealed an unprecedented number of energetically accessible MECIs using a refined algorithm and clustering.
• Demonstrated superior scaling behavior and computational efficiency compared to traditional methods.

Conclusions:

• The parallel on-the-fly algorithm provides a powerful and efficient tool for automated global exploration of CI seam spaces.
• This method significantly advances the ability to discover critical MECIs in complex molecular systems.
• The findings have implications for material and biomedical sciences, particularly in the study of molecular photoswitches.