Answers are grounded in this paper and cite their sources.
Demonstrates a scalable method for calculating ionization energies in large molecules, indicating practical applications.
High Resolution Image Download MS PowerPoint Slide A scalable framework is introduced for the calculation of valence and core ionization energies within the second-order algebraic-diagrammatic construction [ADC(2)] formalism. The approach is based on the construction of state-specific orbital domains that are determined entirely by the underlying electronic structure and ionization process. As a result, the procedure adapts automatically to the character of the ionized state and can be applied in a genuine black-box manner without system-specific tuning. The methodology is tested for conventional ADC(2), its spin-opposite-scaled variant, and an ADC(2)-based double-hybrid functional. Benchmark calculations show that the errors introduced by the local approximation remain far below the intrinsic uncertainties of the underlying correlated methods. For both valence and core ionization energies, the deviations are typically on the order of a few hundredths of an electronvolt. At the same time, substantial reductions of the orbital space are achieved, leading to the significant acceleration of the most expensive steps of the correlated treatment. The efficiency and robustness of the approach are demonstrated for extended molecular systems of practical relevance. Once the reference orbital set is obtained, the valence ionization energy of a 132-atom thermally activated delayed fluorescence emitter can be determined within approximately 20 min using a triple-ζ basis set, while the 4 N K -edge core ionization energies of a 372-atom porphyrin derivative are obtained within about 2 h. In both cases, the corresponding ADC(2) eigenvalue problem itself requires only about 1 min per state. The proposed framework therefore enables routine applications of ADC(2)-based methods to molecular systems that are beyond the reach of conventional implementations.
No takes yet. Share an insight, caveat, or question.
Mester et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: