Artificial Neural Network-Group Contribution Method for Predicting Standard Enthalpy of Formation in the Solid State: C–H, C–H–O, C–H–N, and C–H–N–O Compounds
Abstract:In this work, an artificial neural network-group contribution model is developed to predict the standard enthalpy of formation in the solid (crystal) state of pure compounds. Several classes of hydrocarbon compounds CH, oxygenated compounds CHO, nitrogen compounds CHN, and energetic compounds CHNO are investigated to propose a comprehensive and predictive model. The new model is developed and tested for 1222 organic compounds containing complex molecular structures. The performance of the new model has been co… Show more
“…The fragmentation of the molecular structures was performed by an automatic program using Marrero and Gani method [32]. For more details on the cutting technique we recommend these works [32,37,38] . The chemical structures of all considered compounds were used.…”
Section: Model Developmentmentioning
confidence: 99%
“…For this purpose, in this work the data set was divided into two sub data sets using semi-randomly choice [37,38]. Indeed, if a compound is selected and if this compound is described by one of the irrelevant groups; group having their contribution value generated by less than three compounds, this compound is removed from the test set.…”
Section: Model Developmentmentioning
confidence: 99%
“…After analyzing the molecular structures of these compounds having large estimation errors, new structural groups are defined and included in the Marrero-Gani's groups as additional new third-order groups in order to increase the accuracy and reliability of the proposed model. These new groups provide more structural information and allow distinguishing between similar compounds having one or more functional first-order groups in common [37,38]. A list of new groups, along with sample assignments and group occurrences, is presented in appendix C (See supplementary material).…”
Recently, with the development of calculators and numerical tools, quantum computations to explore the electronic, structural and dynamic properties of matter without resorting to experimental knowledge have seen increasing development. Thus, it is possible to perform ab-initio calculations with increasing precision and for increasingly larger systems. In the scientific literature, papers using ab-initio quantum computation for the prediction of formation enthalpies is more and more numerous. The aim of this paper is to develop a theoretical method to calculate standard enthalpy of formation in gas stat for organic compounds using group contribution technics (third-order group contribution method). For the establishment of this method, 750 molecules are used. In parallel with group contribution methods, this paper presents another approach to calculate gas-state formation enthalpies based on DFT method. The calculation involved 30 molecules with at least one ring from C3 to C13. Finally, DFT and group contribution results are compared.
“…The fragmentation of the molecular structures was performed by an automatic program using Marrero and Gani method [32]. For more details on the cutting technique we recommend these works [32,37,38] . The chemical structures of all considered compounds were used.…”
Section: Model Developmentmentioning
confidence: 99%
“…For this purpose, in this work the data set was divided into two sub data sets using semi-randomly choice [37,38]. Indeed, if a compound is selected and if this compound is described by one of the irrelevant groups; group having their contribution value generated by less than three compounds, this compound is removed from the test set.…”
Section: Model Developmentmentioning
confidence: 99%
“…After analyzing the molecular structures of these compounds having large estimation errors, new structural groups are defined and included in the Marrero-Gani's groups as additional new third-order groups in order to increase the accuracy and reliability of the proposed model. These new groups provide more structural information and allow distinguishing between similar compounds having one or more functional first-order groups in common [37,38]. A list of new groups, along with sample assignments and group occurrences, is presented in appendix C (See supplementary material).…”
Recently, with the development of calculators and numerical tools, quantum computations to explore the electronic, structural and dynamic properties of matter without resorting to experimental knowledge have seen increasing development. Thus, it is possible to perform ab-initio calculations with increasing precision and for increasingly larger systems. In the scientific literature, papers using ab-initio quantum computation for the prediction of formation enthalpies is more and more numerous. The aim of this paper is to develop a theoretical method to calculate standard enthalpy of formation in gas stat for organic compounds using group contribution technics (third-order group contribution method). For the establishment of this method, 750 molecules are used. In parallel with group contribution methods, this paper presents another approach to calculate gas-state formation enthalpies based on DFT method. The calculation involved 30 molecules with at least one ring from C3 to C13. Finally, DFT and group contribution results are compared.
“…where Q detH 2 O-CO 2 is the predicted heat of detonation by the "H 2 O-CO 2 arbitrary", D f H 0 [H 2 O(l)] and D f H 0 [CO 2 (g)] is the standard heat of formation of H 2 O(l) and CO 2 (g), respectively,a nd D f H 0 (explosive) is the condensed phaseh eat of formation of desired explosive. For those energetic compoundsw here their experimental data of the condensed phase heat of formation are not available, simple empirical [3,[35][36][37][38][39],q uantum mechanical [40][41][42][43],g roup additivity [44,45],a nd artificial neural network [46] methods can be used to find the condensedp hase heat of formation of different classes C a H b N c O d energetic compounds.…”
Section: The Use Of the Calculated Heats Of Detonation For Estimationmentioning
The explosive power or strength of an energetic material shows its capacity for doing useful work. This work reviews recent developments for prediction of power of energetic compounds. A new user‐friendly computer code is also introduced to predict the relative power of a desired energetic compound as compared to 2,4,6‐trinitrotoluene (TNT). It is based on the best available methods, which can be used for different types of energetic compounds including nitroaromatics, nitroaliphatics, nitramines, and nitrate esters. The computed relative powers are consistent with the measured data for some new materials containing complex molecular structures.
“…It is a valuable indicator of crystal stability and may be used to evaluate other properties of practical significance such as vapor pressure . Moreover, it is a significant contribution to the solid-state formation enthalpy Δ f H 0 (cr) of organic crystals required to evaluate the energy content, performances, and sensitivities of candidate materials of potential interest as components of fuels, propellants, or explosives. , Although Δ f H 0 (cr) may be estimated directly from molecular structure, i.e., without explicitly evaluating the gas-phase and sublimation enthalpies, such procedures are deprived of physical bases and rely mostly on empiricism. − Since the intra- and intermolecular contributions to Δ f H 0 (cr) depend on distinct interactions, they should ideally be evaluated separately.…”
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.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.