Accurate prediction of heats of formation by a combined method of B3LYP and neural network correction

Toward Chemical Accuracy in Predicting Enthalpies of Formation with General-Purpose Data-Driven Methods

Enthalpies of formation and reaction are important thermodynamic properties that have a crucial impact on the outcome of chemical transformations. Here we implement the calculation of enthalpies of formation with a general-purpose ANI-1ccx neural network atomistic potential. We demonstrate on a wide range of benchmark sets that both ANI-1ccx and our other general-purpose data-driven method AIQM1 approach the coveted chemical accuracy of 1 kcal/mol with the speed of semiempirical quantum mechanical methods (AIQM1) or faster (ANI-1ccx). It is remarkably achieved without specifically training the machine learning parts of ANI-1ccx or AIQM1 on formation enthalpies. Importantly, we show that these data-driven methods provide statistical means for uncertainty quantification of their predictions, which we use to detect and eliminate outliers and revise reference experimental data. Uncertainty quantification may also help in the systematic improvement of such data-driven methods.

Learning Molecular Representations for Thermochemistry Prediction of Cyclic Hydrocarbons and Oxygenates

Accurate thermochemistry estimation of polycyclic molecules is crucial for kinetic modeling of chemical processes that use renewable and alternative feedstocks. In kinetic model generators, molecular properties are estimated rapidly with group additivity, but this method is known to have limitations for polycyclic structures. This issue has been resolved in our work by combining a geometry-based molecular representation with a deep neural network trained on ab initio data. Each molecule is transformed into a probabilistic vector from its interatomic distances, bond angles, and dihedral angles. The model is tested on a small experimental dataset (200 molecules) from the literature, a new medium-sized set (4000 molecules) with both open-shell and closed-shell species, calculated at the CBS-QB3 level with empirical corrections, and a large G4MP2-level QM9-based dataset (40 000 molecules). Heat capacities between 298.15 and 2500 K are calculated in the medium set with an average deviation of about 1.5 J mol^-1 K^-1 and the standard entropy at 298.15 K is predicted with an average error below 4 J mol^-1 K^-1. The standard enthalpy of formation at 298.15 K has an average out-of-sample error below 4 kJ mol^-1 on a QM9 training set size of around 15 000 molecules. By fitting NASA polynomials, the enthalpy of formation at higher temperatures can be calculated with the same accuracy as the standard enthalpy of formation. Uncertainty quantification by means of the ensemble standard deviation is included to indicate when molecules that are on the edge or outside of the application range of the model are evaluated.

Accurate heats of formation of polycyclic saturated hydrocarbons predicted by using the XYG3 type of doubly hybrid functionals

Polycyclic saturated hydrocarbons (PSHs) are attractive candidates as hydrocarbon propellants. To assess their potential values, one of the key factors is to determine their energy contents, such as to calculate their heats of formation (HOF). In this work, we have calculated HOFs for a set of 36 PSHs including exo-Tricyclo[5.2.1.0^(2,6) ] decane, the principal component of the high-energy density hydrocarbon fuel commonly identified as JP-10. The results from B3LYP, B3LYP-D3BJ, M06-2X, B2PLYP, B2PLYP-D3BJ, and the XYG3 type of doubly hybrid (xDH) functionals are presented. It is demonstrated here that the xDH functionals yield accurate HOFs in good agreement with those from experiments or the G4 theory. In particular, XYGJ-OS, a low scaling xDH functional, is shown to hold the promise for accurate prediction of HOFs for PSHs of larger sizes. © 2018 Wiley Periodicals, Inc.

Beyond energies: geometry predictions with the XYG3 type of doubly hybrid density functionals

The scaled mean absolute deviations (s-MADs) of the optimized geometric parameters for covalent bondings (the CCse set), nonbonded interactions (the S22G30 set) and the transition state structures (the TSG36 set), with Tot referring to the averaged s-MAD for general performances.