Efficient partitioning technique for computing the dynamics of intramolecular processes: radiationless transitions in...

Area of Science:

• Quantum Chemistry
• Computational Physics
• Molecular Dynamics

Background:

• Accurate computation of molecular system dynamics is crucial for understanding chemical reactions.
• Traditional methods like diagonalization face scalability challenges with increasing system size.

Purpose of the Study:

• To present an efficient Quantum Partitioning (QP) algorithm for computing eigenvalues, eigenvectors, and dynamics of large molecular systems.
• To demonstrate the algorithm's advantages in terms of computational scaling, parallelization, and memory requirements.

Main Methods:

• Developed an efficient QP partitioning algorithm.
• Implemented a coarse-grained approximation to enhance basis set size.
• Applied the algorithm to simulate electronic internal conversion dynamics in a pyrazine model.

Main Results:

• The QP algorithm exhibits favorable N(T)^2 scaling, outperforming straightforward diagonalization.
• The method is trivially parallelizable, achieving linear speedup.
• The coarse-grained approximation allows for unprecedented basis set sizes (10^8-10^10).
• Simulations of pyrazine decay dynamics and resonance line shapes were successfully performed.

Conclusions:

• The QP partitioning algorithm provides an efficient and scalable approach for large molecular system dynamics.
• The method's parallelizability and modest storage requirements make it suitable for high-performance computing.
• The enhanced algorithm opens new possibilities for simulating complex molecular processes.