Accelerating Realtime TDDFT with Block-Orthogonalized Manby-Miller Embedding Theory.

Area of Science:

• Computational chemistry
• Quantum mechanics
• Materials science

Background:

• Realtime time-dependent density-functional theory (RT-TDDFT) is crucial for simulating molecular electronic dynamics.
• Simulating large, solvated systems over long timescales with RT-TDDFT is computationally expensive.
• Accelerating RT-TDDFT is essential for applications in nonlinear spectroscopy and energy transport.

Purpose of the Study:

• To adapt an existing embedding technique for accelerating RT-TDDFT simulations.
• To evaluate the accuracy and efficiency of the proposed embedding method.
• To compare the embedded mean-field theory scheme with QM/MM charge embeddings.

Main Methods:

• Application of Manby and Miller's embedding technique to RT-TDDFT.
• Assessment of accuracy and speed using absorption spectra of solvated and covalently split chromophores.
• Comparison with quantum mechanics/molecular mechanics (QM/MM) charge embedding methods.

Main Results:

• The embedded mean-field theory (EMFT) scheme effectively accelerates RT-TDDFT simulations.
• The EMFT approach demonstrates good accuracy in predicting absorption spectra.
• Mixing levels of detail in the EMFT scheme provides a balance between accuracy and computational cost.

Conclusions:

• The embedded mean-field theory scheme is a simple, accurate, and effective method for accelerating RT-TDDFT.
• This technique enables more efficient simulations of large solvated systems for electronic dynamics.
• The approach holds promise for advancing studies in nonlinear spectroscopy and energy transport.