PC-SAFT Parameter Predictions for Ionic Liquids and Deep Eutectic Solvents

A Graph Neural Network model predicts Perturbed-Chain Statistical Associating Fluid Theory (PC-SAFT) parameters for 1842 molecules based on molecular structure. The model, trained on experimental vapor pressure and saturated liquid density data, demonstrated superior performance on a test set of 1642 unseen molecules and a subset of 122 molecules compared to a traditional group contribution method. Created by Wildson Bernardino de Brito Lima and last updated on 2026-05-28, the dataset supports the application of the PC-SAFT equation of state to complex, low-volatility solvents.

Use Cases

• Predicting PC-SAFT pure-component parameters for novel molecules based on their molecular structure.
• Estimating liquid density and vapor pressure for ionic liquids lacking experimental data.
• Parametrizing deep eutectic solvents for thermodynamic property modeling.
• Comparing the performance of graph neural network-based parameter prediction against traditional group contribution methods.

Strengths

• Model trained on a dataset of 1842 molecules and evaluated on a test set of 1642 unseen molecules.
• Achieved a liquid density mean absolute percentage error (MAPE) of 8.21% and a vapor pressure mean absolute log error (MALE) of 0.4271 on the test set.
• For a compatible subset of 122 molecules, reduced density MAPE from 7.17% to 3.79% and vapor pressure MALE from 1.3470 to 0.2829 compared to GC-PCSAFT.
• Successfully parametrized complex solvents, yielding low density MAPEs (e.g., 0.20% for [Bmim][Tf2N]) and evaluated 24 other deep eutectic solvents.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset is small in scale at 799.2 KB.

Provenance

Source

figshare

Collection Method

Parameters derived from experimental vapor pressure and saturated liquid density data using a Graph Neural Network model.

Freshness

Last updated 2026-05-28 06:14:06; freshness should be verified.