The third-order algebraic diagrammatic construction method (ADC(3)) for the polarization propagator for closed-shell...

Area of Science:

• Computational Chemistry
• Quantum Chemistry
• Theoretical Chemistry

Background:

• Accurate computation of molecular excited states is crucial for understanding photophysical and photochemical processes.
• Perturbation theory methods, such as algebraic diagrammatic construction (ADC), offer a balance between accuracy and computational cost.
• Previous studies have established benchmark datasets for evaluating the performance of quantum chemical methods for excitation energies.

Purpose of the Study:

• To report an efficient implementation of the algebraic diagrammatic construction method in third-order perturbation theory (ADC(3)) for calculating excited states.
• To assess the accuracy of ADC(2) and ADC(3) methods by comparing their results with a recently established benchmark set.
• To compare the performance of ADC(3) with the coupled cluster singles and doubles in triples (CC3) method.

Main Methods:

• Implementation of an efficient computational program for the ADC(3) method.
• Calculation of vertical excited singlet and triplet states for 28 organic molecules.
• Comparison of calculated excitation energies and oscillator strengths with Thiel's benchmark dataset and CC3 results.

Main Results:

• ADC(3) demonstrated high accuracy with mean errors of 0.12 ± 0.28 eV (singlets) and -0.18 ± 0.16 eV (triplets) against theoretical best estimates.
• ADC(2) variants showed varying accuracies, with ADC(2)-x performing better than ADC(2)-s for triplet states.
• Comparison with CC3 indicated comparable accuracy for ADC(3), with ADC(3) exhibiting slightly better performance for singlet states (0.08 ± 0.27 eV vs 0.23 ± 0.21 eV).

Conclusions:

• The ADC(3) method provides a highly accurate and efficient approach for computing excited states in organic molecules.
• ADC(3) demonstrates a performance comparable to the more computationally expensive CC3 method.
• ADC(3) possesses a significantly larger range of applicability due to its favorable O(N^6) scaling with system size, making it suitable for larger molecular systems.