Predicting Excited-State Energies from Ground-State Descriptors in Thermally Fluctuating π-Conjugated Molecules

Area of Science:

• Computational chemistry
• Materials science
• Organic electronics

Background:

• Predicting excitation energies in organic semiconductors is crucial for designing new materials but computationally expensive.
• High-accuracy excited-state methods limit extensive thermal sampling for accurate energy predictions.

Purpose of the Study:

• To develop an efficient and interpretable regression framework for predicting excitation energies in organic semiconductors.
• To enable accurate thermal sampling and understanding of excited-state energetics in these materials.

Main Methods:

• Developed a regression framework using ground-state electronic descriptors (HOMO-LUMO gap and exchange integral).
• Validated the model on thermally sampled configurations of tetracene derivatives and related π-conjugated systems.
• Demonstrated backbone-level transferability for derivatives beyond tetracene.

Main Results:

• The framework accurately predicts singlet and triplet excitation energies, correlating them with HOMO-LUMO gap and exchange integral.
• Model predictions for E(S1) distribution maxima align with experimental absorption maxima.
• The approach captures excitation energy distributions due to structural fluctuations.

Conclusions:

• The descriptor-based mapping provides an efficient and physically transparent method for describing thermally modulated excited-state energetics.
• This framework aids in the design of excitonic materials and organic semiconductors.
• The model shows transferability across different π-conjugated systems and backbones.