← Benchmarks | Solvation

Solvation Thermodynamics BENCHMARK

Scope: Aqueous solubility validated on 1,128 ESOL compounds. Four-term solvation model from first principles. Full methodology published.

0.71
logS MAE
ESOL (1,128 compounds)
1,128
Compounds
Delaney ESOL dataset
30+
Solvents
Supported in library
~10ms
Per Molecule
Prediction time

Methodology

How FluxMateria predicts solvation properties

FLUX Four-Term Solvation Model

Solvation free energies are computed from four physically-motivated terms, each derived from molecular geometry. No training data required.

1. Cavity Formation

Energy to create a solute-sized cavity in the solvent. Scales with molecular surface area.

2. Dispersion

van der Waals interactions between solute and surrounding solvent molecules.

3. Electrostatic

Charge-dipole and dipole-dipole interactions from molecular polarity.

4. Hydrogen Bonding

Directional H-bond donor/acceptor contributions from FLUX geometry.

Validation Approach

ESOL Dataset

1,128 compounds from the Delaney aqueous solubility dataset. Standard benchmark for solubility prediction models. All experimental logS values.

AqSolDB Cross-Check

Additional validation on AqSolDB (0.84 logS MAE), confirming generalization beyond the primary ESOL test set.

Aqueous Solubility Results

Performance on standard benchmark datasets

Dataset Compounds logS MAE RMSE Bias Status
ESOL (Delaney) 1,128 0.71 0.913 +0.024 PASS
AqSolDB 9,982 0.84 1.08 -0.11 PASS

ESOL target: MAE < 0.8 logS units. Both datasets pass the commercial-grade accuracy threshold.

Comparison with Established Methods

How FluxMateria compares to solvation models and ML approaches

Method logS MAE Speed Training Data Parameters
FluxMateria 0.71 ~10 ms None 0 fitted
ESOL Consensus Model 0.75 ~1 ms Fitted descriptors 4 fitted
Random Forest (fingerprints) 0.69 ~5 ms Training set required Thousands
SMD (Implicit Solvation) 0.6-1.0 kcal/mol* Minutes-hours None Atomic radii fitted
COSMO-RS 0.5-0.8 kcal/mol* Minutes None Several fitted

* SMD and COSMO-RS report hydration free energy (kcal/mol), not logS directly. Values are approximate conversions for comparison purposes. FluxMateria achieves competitive accuracy with instant predictions.

Solvent Coverage

30+ solvents supported across multiple classes

Protic Solvents

Water, methanol, ethanol, isopropanol, n-butanol, acetic acid, formic acid

Aprotic Polar

DMSO, DMF, acetonitrile, acetone, NMP, THF

Nonpolar

Hexane, toluene, benzene, cyclohexane, diethyl ether, DCM, chloroform

Specialized

Carbon tetrachloride, nitromethane, pyridine, 1,4-dioxane, ethyl acetate, MTBE

Physical Consistency Tests

Thermodynamic relationships verified across predictions

  • More polar molecules more soluble in water ✓ 100%
  • Increasing molecular weight decreases aqueous solubility ✓ Correct trend
  • H-bond donors/acceptors increase water solubility ✓ 100%
  • All four solvation terms derive from first principles ✓ Verified

Scope & Limitations

Where predictions are most and least reliable

Strengths

  • Aqueous solubility validated on 1,128 ESOL compounds (MAE 0.71)
  • Cross-validated on AqSolDB (9,982 compounds, MAE 0.84)
  • Competitive with ML methods that require training data
  • Fully reproducible — no retraining required
  • ~10ms per molecule — suitable for high-throughput screening

Known Limitations

  • FreeSolv hydration free energy benchmark configured but not yet fully validated
  • logP (octanol/water) validation in progress
  • Organic solvent solvation supported but benchmark data acquisition ongoing
  • Performance may vary for ionic compounds and highly charged species
  • Mixed solvent systems are planned but not yet implemented

References

Primary data sources for experimental validation

  1. J.S. Delaney, "ESOL: Estimating Aqueous Solubility Directly from Molecular Structure," J. Chem. Inf. Comput. Sci., 2004, 44, 1000-1005.
  2. S. Sorkun, A. Khetan, S. Er, "AqSolDB: A curated reference set of aqueous solubility," Sci. Data, 2019, 6, 143.
  3. D.L. Mobley, J.P. Guthrie, "FreeSolv: A database of experimental and calculated hydration free energies," J. Comput.-Aided Mol. Des., 2014, 28, 711-720.
  4. A.V. Marenich, C.J. Cramer, D.G. Truhlar, "Universal Solvation Model Based on Solute Electron Density (SMD)," J. Phys. Chem. B, 2009, 113, 6378-6396.
  5. A. Klamt, "COSMO-RS: From Quantum Chemistry to Fluid Phase Thermodynamics and Drug Design," Elsevier, 2005.

Try the Solvation module

Predict solvation thermodynamics across multiple solvents with full interpretability.

← Back to Module Request Access