Solubility prediction for ill-defined polymers: A critical comparison of Hansen parameters and COSMO-RS backfitting
Abstract
Selecting solvents for polymeric materials is challenging because the target solute is rarely a single well-defined molecular species, but rather a system with chemical, physical, and formulation complexity. To address this issue, this short review compares two “inverse-method” workflows in which solvent-screening data are used to reconstruct effective descriptors of such ill-defined solutes. Hansen solubility parameters (HSP) provide a pragmatic representation of solubility by translating solvent-screening data into a spherical compatibility domain, but their predictive scope remains constrained by the “like dissolves like” paradigm. COSMO-RS backfitting, a largely overlooked inverse extension of COSMO-RS, uses the same experimental logic to reconstruct an effective σ-profile, enabling solvent ranking in a thermodynamic framework without requiring an explicit polymer structure. Using cellulose triacetate as a stringent benchmark, the review shows that the two approaches are best viewed as sequentially complementary: HSP provides a first operational and visually intuitive map, whereas COSMO-RS backfitting offers greater predictive value for ranking new pure solvents when geometric similarity is no longer sufficiently discriminating.
Related
- Hybrid-CatBoost: an empirical comparison of molecular representations for aqueous-solubility prediction on the TDC benchmark
- Solubility parameters of enzymatic hydrolysis lignin: a comparative study of inverse gas chromatography and Hansen solubility sphere.
- An Inverse Design Workflow for Solvent Mixtures Based on Hansen Solubility Parameters
- Investigating Topological Parameters in Polymers: Precision Synthesis of Trefoil Polymers from a Knotted Synthon.
- Application of Hansen Solubility Parameters for Green Solvent Extraction of Flavonoids from Oroxylum indicum (L.) Kurz and Chromatographic Profiling Across Plant Organs
BibTeX
@article{aubry2026solubility,
title = {{Solubility prediction for ill-defined polymers: A critical comparison of Hansen parameters and COSMO-RS backfitting}},
author = {Jean-Marie Aubry and Evamaria Hofmann and Florian Laubé and Théophile Gaudin and Véronique Nardello-Rataj},
journal = {Current Opinion in Colloid \& Interface Science},
year = {2026},
month = sep,
doi = {10.1016/j.cocis.2026.102044},
url = {https://hal.science/hal-05734333v1},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Solubility prediction for ill-defined polymers: A critical comparison of Hansen parameters and COSMO-RS backfitting
AU - Jean-Marie Aubry
AU - Evamaria Hofmann
AU - Florian Laubé
AU - Théophile Gaudin
AU - Véronique Nardello-Rataj
PY - 2026
DA - 2026/09/01
JO - Current Opinion in Colloid & Interface Science
DO - 10.1016/j.cocis.2026.102044
UR - https://hal.science/hal-05734333v1
AB - Selecting solvents for polymeric materials is challenging because the target solute is rarely a single well-defined molecular species, but rather a system with chemical, physical, and formulation complexity. To address this issue, this short review compares two “inverse-method” workflows in which solvent-screening data are used to reconstruct effective descriptors of such ill-defined solutes. Hansen solubility parameters (HSP) provide a pragmatic representation of solubility by translating solvent-screening data into a spherical compatibility domain, but their predictive scope remains constrained by the “like dissolves like” paradigm. COSMO-RS backfitting, a largely overlooked inverse extension of COSMO-RS, uses the same experimental logic to reconstruct an effective σ-profile, enabling solvent ranking in a thermodynamic framework without requiring an explicit polymer structure. Using cellulose triacetate as a stringent benchmark, the review shows that the two approaches are best viewed as sequentially complementary: HSP provides a first operational and visually intuitive map, whereas COSMO-RS backfitting offers greater predictive value for ranking new pure solvents when geometric similarity is no longer sufficiently discriminating.
SN - 1359-0294
SN - 1879-0399
ER -