These are preliminary reports that have not been peer-reviewed. They should not be regarded as conclusive, guide clinical practice/health-related behavior, or be reported in news media as established information. For more information, please see our FAQs.
3 files

Predicting Carbonyl Excitation Energies Efficiently Using EOM-CC Trends

revised on 11.09.2020, 07:45 and posted on 11.09.2020, 08:50 by Keiran Rowell, Scott Kable, Meredith J. T. Jordan
We approach the problem of predicting excitation energies of diverse, larger (5–6 carbons) carbonyl species central to earth’s tropospheric chemistry. Triples contributions are needed for the vertical excitation energy (Evert), while EOM-CCSD//TD-DFT calculations provide acceptable estimates for the S1 relaxation energy (Erelax), and (TD-)DFT suffices for the S0 → S1 zero-point vibration energy correction (∆EZPVE).

Perturbative triples corrections deliver Evert values close in accuracy to full iterative triples EOM-CC calculations. The error between EOM-CCSD and triples-corrected E vert values appears to be systematic and can be accounted for with scaling factors. However, saturated and α,β-unsaturated carbonyls must be treated separately. Double-hybrid S0 minima can be used to calculate Evert with negligible loss in accuracy, relegating the O(N5) expense of CCSD to only single-point energy and excitation calculations.

This affordable protocol can be applied to all volatile carbonyl species. E0−0 predictions do overestimate measured values by ∼8 kJ/mol due to a lack of triples contribution in E relax, but this overestimation is systematic and the mean unsigned error is within 4 kJ/mol once this is accounted for.


Atmospheric photochemistry

Australian Research Council

Find out more...


Email Address of Submitting Author


The University of Sydney



ORCID For Submitting Author


Declaration of Conflict of Interest

No conflict of interest

Version Notes

2nd version: -Updated 0-0 best estimate table to include MSE error shift for small carbonyls, and correct dZPVE for large carbonyls. Previously dZPVE in this best estimates table was mistakenly ZPVE(TS) - ZPVE(min), not ZPVE(S1 min) - ZPVE(S0 min).


Logo branding