New empirical correlations for the prediction of critical properties and acentric factor of S-containing compounds

Ghomisheh, Zahra; Sobati, Mohammad Amin; Gorji, Ali Ebrahimpoor

doi:10.1080/17415993.2021.2017936

Cited by 2 publications

(5 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are a total of eight targets to predict, four of which are the critical properties and acentric factor (main targets) and the other four are phase change properties (auxiliary targets). Several studies indicate that multitask learning with related targets can improve generalization and prediction accuracy. , Many empirical correlations imply the underlying link among T c , T b , P c , ρ c , and ω , and therefore, multitarget training is expected to help model predictions. Several target groupings are explored to optimize model performance.…”

Section: Methodsmodelsmentioning

confidence: 99%

“…Several studies indicate that multitask learning with related targets can improve generalization and prediction accuracy. 41,42 Many empirical correlations imply the underlying link among T c , T b , P c , ρ c , and ω 43,44 and therefore, multitarget training is expected to help model predictions. Several target groupings are explored to optimize model performance.…”

Section: Multitask Learning and Targetmentioning

confidence: 99%

“…35,37,39−42 The ML model for critical properties and acentric factor prediction is expected to particularly benefit from multitask learning as correlations relating T c , P c , ρ c , and ω, with boiling point (T b ), have been established in literature. 43,44 In this work, we construct multitask models with a graph convolution neural network (GCNN) to predict the three critical properties (T c , P c , ρ c ) and acentric factor (ω). Directed message passing neural networks (D-MPNN) 32 and graph attention networks (GAT) 45 are explored as the GCNN layers, which can learn an optimized latent space molecular representation during the prediction tasks.…”

Section: Introductionmentioning

confidence: 99%

“…The underlying assumption of multitask learning is that it can leverage knowledge shared by other related targets and learn more generalized representations. The advantage of a multitask approach has been demonstrated in many applications including drug discovery and bioinformatics. ,,− The ML model for critical properties and acentric factor prediction is expected to particularly benefit from multitask learning as correlations relating T c , P c , ρ c , and ω, with boiling point ( T b ), have been established in literature. , …”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Predicting Critical Properties and Acentric Factors of Fluids Using Multitask Machine Learning

Biswas

Chung

Ramírez

et al. 2023

J. Chem. Inf. Model.

View full text Add to dashboard Cite

Knowledge of critical properties, such as critical temperature, pressure, density, as well as acentric factor, is essential to calculate thermo-physical properties of chemical compounds. Experiments to determine critical properties and acentric factors are expensive and time intensive; therefore, we developed a machine learning (ML) model that can predict these molecular properties given the SMILES representation of a chemical species. We explored directed message passing neural network (D-MPNN) and graph attention network as ML architecture choices. Additionally, we investigated featurization with additional atomic and molecular features, multitask training, and pretraining using estimated data to optimize model performance. Our final model utilizes a D-MPNN layer to learn the molecular representation and is supplemented by Abraham parameters. A multitask training scheme was used to train a single model to predict all the critical properties and acentric factors along with boiling point, melting point, enthalpy of vaporization, and enthalpy of fusion. The model was evaluated on both random and scaffold splits where it shows state-of-the-art accuracies. The extensive data set of critical properties and acentric factors contains 1144 chemical compounds and is made available in the public domain together with the source code that can be used for further exploration.

show abstract

Section: Methodsmodelsmentioning

confidence: 99%

Section: Multitask Learning and Targetmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Predicting Critical Properties and Acentric Factors of Fluids Using Multitask Machine Learning

Biswas

Chung

Ramírez

et al. 2023

J. Chem. Inf. Model.

View full text Add to dashboard Cite

show abstract

“…A new second order group contribution method has been developed [38] to predict the acentric factors of organic compounds. Using the normal boiling temperature, molecular weight and the number of atoms and bonds, empirical correlations are developed [39] to estimate the acentric factors of s-containing compounds. Two different intelligent systems are used to estimate [40] the acentric factors of binary and ternary mixtures of ionic liquids.…”

Section: Introductionmentioning

confidence: 99%

Temperature dependence of the acentric factor of normal hydrogen, orthohydrogen and parahydrogen

Ramesh

Balasubramanian

2022

Preprint

View full text Add to dashboard Cite

Temperature-dependence correlations of vapor pressure and acentric factor for normalhydrogen(n-H2), orthohydrogen(o-H2) and parahydrogen(p-H2), have been formulated. The obtained correlations are statistically excellent. The characteristic parameters such as the Pitzer’s acentric factor, Riedel’s parameter, Filippov’s parameter have been determined for n-H2, o-H2 and p-H2. And, the curvatures of vapor pressure curve for n-H2, o-H2 and p-H2 have been determined in a wide range of temperature. It is found that the curvatures of vapor pressure curve for n-H2, o-H2 and p-H2 have a maximum at about 17.11K, 17.12K and 17.00K, respectively.

show abstract

New empirical correlations for the prediction of critical properties and acentric factor of S-containing compounds

Cited by 2 publications

References 47 publications

Predicting Critical Properties and Acentric Factors of Fluids Using Multitask Machine Learning

Predicting Critical Properties and Acentric Factors of Fluids Using Multitask Machine Learning

Temperature dependence of the acentric factor of normal hydrogen, orthohydrogen and parahydrogen

Contact Info

Product

Resources

About