2018 AIChE Annual Meeting
(377r) Predictions of Gas Phase Thermochemical Properties from Ab Initio Calculation: Applications to Bio-Oil Compounds
Authors
In this work, we extend the work of Paulechka and Kazakov [3] and propose a general predictive approach based on the very accurate and efficient DLPNO-CCSD(T) ab initio calculation, to determine the enthalpies of formation of broad variety of molecules. New sets of regressed atomic contributions are proposed for a larger number of elements: H, C, N, O, F, Si, P, S, Cl and Br, and different ab initio methods for the calculation of the electronic energies and the determination of the optimal molecular geometry. Excellent predictions are obtained for most studied compounds (bio-oil compounds, refrigerants, â¦), apart from certain types of compounds for which the experimental measurements can be discussed. The predictive approach appears to be insensitive to the ab initio method used, given the very different performance of the employed DFT-functional in benchmark studies, and seems to be a very promising method for the prediction of the formation enthalpy of species that are difficult to measure or where significant uncertainty in the measurement exists. The combination of DFT and DLPNO-CCSD(T) with the def2/J and def2-QZVPP/C auxiliary basis sets and the RIJCOSX approximation provide a computationally very efficient low cost path for the calculation of formation enthalpies.
[1] A. Osmont, L. Catoire, I. Gökalp, and V. Yang, âAb initio quantum chemical predictions of
enthalpies of formation, heat capacities, and entropies of gas-phase energetic compounds,â
Combustion and Flame, 151, 262 â 273 (2007).
[2] A. Demenay, J. Glorian, P. Paricaud, and L. Catoire, âPredictions of the ideal gas properties
of refrigerant molecules,â International Journal of Refrigeration, 79, 207 â 216 (2017).
[3] E. Paulechka and A. Kazakov, âEfficient DLPNOâCCSD(T)-based estimation of formation
enthalpies for C-, H-, O-, and N-containing closed-shell compounds validated against critically
evaluated experimental data,â J. Phys. Chem. A, 121, 4379â4387 (2017)