Ab initio nonadiabatic dynamics of multichromophore complexes: a scalable graphical-processing-unit-accelerated...

Area of Science:

• Computational chemistry and physics
• Materials science
• Quantum mechanics

Background:

• Molecular simulations are advancing due to hardware and algorithm improvements.
• Challenges remain for complex chemical systems like electrochemistry and condensed phase reactivity.
• Fragment-based approaches offer a promising strategy for parallel architectures.

Purpose of the Study:

• To develop new methods for large-scale molecular simulations.
• To address limitations in simulating complex chemical systems.
• To exploit emerging parallel computing architectures.

Main Methods:

• Utilizing fragment-based approaches for electronic structure theory.
• Employing graphical processing units (GPUs) for accelerated calculations.
• Developing an exciton model to stitch together solutions for fragmented systems.
• Implementing a multitiered parallel framework for nonadiabatic dynamics simulations.

Main Results:

• Fragment-based methods show significant benefits, especially for computationally intensive calculations.
• GPU-accelerated electronic structure theory calculations were advanced.
• A parallel excitonic framework was developed for large multichromophoric assemblies.
• The framework demonstrated good agreement with computationally demanding TDDFT calculations.

Conclusions:

• The developed multitiered parallel framework enables simulations of large systems and long timescales.
• This approach is well-suited for emerging parallel computing trends.
• Advances in fragment-based methods and GPU acceleration are crucial for future molecular simulations.