Measurements and Prediction of the Partitioning of Volatile Organic Compounds between Air and Cotton

The cotton-air partition coefficient (K_ca) is a key parameter governing the concentration and human exposure to volatile organic compounds (VOCs) in indoor environments. We experimentally determined K_ca for 46 diverse VOCs, including carbonyls, aromatics, amines, amides, alkenes, acids, alcohols, furan, and siloxanes. Results showed that compounds with a lower vapor pressure and higher molecular weight exhibited stronger sorption. Water sorbed in cotton contributed less than half an order of magnitude to K_ca for most VOCs, except for some highly water-soluble species. Octanol-air partition coefficients (log K_oa) correlate well with log K_ca for homologous compounds but demonstrate limited predictive capacity across different VOC classes. The predictions of log K_ca using methyl cellulose as a cotton surrogate led to deviations of up to one log unit. We developed a polyparameter linear free energy relationship (pp-LFER) model for log K_ca prediction based on the experimental data. The model shows good agreement with experimental data (adjusted R² = 0.73, root-mean-square error ∼ 0.55 log units), providing a useful tool for chemical distribution and exposure assessment. Field investigations further demonstrated that environmental factors (temperature, humidity, particle deposition, and interfering compounds) can induce 1-2 orders of magnitude variation in K_ca values compared to controlled laboratory measurements.