Accurate or Fast Prediction of Solid-State Formation Enthalpies Using Standard Sublimation Enthalpies Derived From Geometrical Fragments

Abstract

In view of its success in predicting sublimation enthalpies of molecular crystals near triple point conditions [Ind. Eng. Chem. Res. 2012, 51, 2814], the geometrical fragment (GF) approach is presently used to estimate standard values Δsub H 0 adjusted to the reference temperature of 298.15 K. The resulting GF0 scheme is fitted against theoretically confirmed Δsub H 0 data for 185 high nitrogen compounds and validated using organic compounds not used in the fitting process. It is subsequently used to predict the solid-state formation enthalpy Δf H 0(cr) of 40 energetic materials, combined with either correlated ab initio methods or simple models to estimate the gas-phase contribution. The combination of GF0 with ab initio data yields Δf H 0(cr) values with state-of-the-art accuracy. Combined with the simple atom pair contribution model, this procedure predicts Δf H 0(cr) with better accuracy and stronger physical grounds than alternative low-cost methods.

Machine-Learning Assisted Screening of Energetic Materials

“…For a given material candidate, the heat of formation Δ H f is calculated by the atom pair contribution (APC) method. , This method provides both computational efficiency similar to the group additive method and accuracy close to quantum mechanics under most circumstances.…”

“…(1) For a given material candidate, the heat of formation ΔH f is calculated by the atom pair contribution (APC) method. 73,74 This method provides both computational efficiency similar to the group additive method and accuracy close to quantum mechanics under most circumstances. (2) After determining the OB of given C a H b N c O d candidate by eq 4, the corresponding decomposition reaction is estimated by the K−W rules for OB larger than 40% and the modified K−W rules for OB smaller than 40%.…”

Molecular Energies Derived from Deep Learning: Application to the Prediction of Formation Enthalpies Up to High Energy Compounds

“…Therefore, the prediction of the performances of any EM involves evaluating its enthalpy of formation. This is most rigorously done by subtracting the standard enthalpy of sublimation of the crystal from the enthalpy of formation of the molecule in gas phase [2] . The amount of experimental studies needed to develop new EMs directly depends on the confidence that can be placed in property estimation methods, hence on the accuracy of predicted formation enthalpies.…”

“…Such results are very promising regarding the potential of machine learning (ML) for the prediction of thermochemical properties of gas‐phase molecules, especially in view of predicting from ML results instead of resorting to costly QC methods. Associated with an efficient procedure to predict sublimation enthalpies, [2] machine learned data should provide a way to the fast prediction of solid‐state formation enthalpies. This would be extremely valuable in the field of energetic materials, for which machine learning approaches are increasingly used to search for compounds with desirable properties [14,15] .…”

“…This is most rigorously done by subtracting the standard enthalpy of sublimation of the crystal D sub H 0 from the enthalpy of formation D f H 0 of the molecule in gas phase. [2] The amount of experimental studies needed to develop new EMs directly depends on the confidence that can be placed in property estimation methods, hence on the accuracy of predicted formation enthalpies. In the last two decades, high levels of quantum chemical theory have been increasingly used to compute D f H 0 with the highest possible accuracy.…”