(257j) Thermodynamic Solubility Modeling of 2,4,6-Trinitrotoluene (TNT) | AIChE

(257j) Thermodynamic Solubility Modeling of 2,4,6-Trinitrotoluene (TNT)


Hossain, N. - Presenter, Texas Tech University
Bhattacharia, S. - Presenter, Texas Tech University
Chen, C. C. - Presenter, Texas Tech University

Thermodynamic solubility modeling of energetic materials in solvents and polymer binders is a key enabling technology for development of novel, safe and efficient process technology and formulations for these energetics materials, and to reliably predict their environmental fate. However, few thermodynamic studies on solubilities of energetic materials in solvents have been reported in the literature. Recent success in the solubility modeling of pharmaceutical molecules with the Non-Random Two-Liquids Segment Activity Coefficient model (NRTL-SAC) [Tung, H.-H., Org. Process Res. Dev., 17, 445-454 (2013); Tanveer, S., Y.-F. Hao, C.-C. Chen, Chem. Eng. Prog., 110 (9), 37-47 (2014)] shows effective solubility modeling methodology should involve both the correlative power of the model to identify molecular parameters from trusted data for solubility in representative solvents and the predictive power of the model to predict solubility in other solvents and solvent mixtures. In this work, we show accurate solubility modeling for 2,4,6-trinitrotoluene (TNT) in solvents including water, ethanol, CCl4, chloroform, acetone, and pyridine and a binary solvent of  methanol and water. The segment parameters of TNT for NRTL-SAC model are first regressed from experimental TNT solubility data in selected single solvents and then they are used to predict TNT solubility in other single solvents and solvent mixtures. We show NRTL-SAC provides semi-quantitative representation for the solubility of TNT in various solvents and solvent mixtures while the predictive UNIFAC and COSMO-SAC models provide qualitative representation of the TNT solubility data.