Costa G.P., Choi P., Stoyanov S.R., Liu Q. (2024) Journal of Molecular Modeling, 30 (7), art. no. 196,   DOI: 10.1007/s00894-024-05929-w

ABSTRACT: Context: To determine the miscibility of liquids at high temperatures using the concept of Hildebrand solubility parameter δ, the current practice is to examine the difference in δ between two liquids at room temperature, assuming that δ is not sensitive to temperature. However, such an assumption may not be valid for certain polymer blends and solutions. Therefore, a knowledge of the δ values of the liquids of interest at high temperatures is desirable. The determination of δ at high temperatures, especially for high-molecular-weight polymers, is impossible, as polymers have vapor pressures of zero. To this end, molecular dynamics (MD) simulations provide a practical means for determining δ over a wide range of temperatures. In this work, we study the temperature dependence of δ of five hydrocarbon polymers: polyethylene (PE), isotactic and atactic polypropylene (i-PP and a-PP), polyisobutylene (PIB), and polyisoprene (PI) in five hydrocarbon solvents: n-pentane, n-hexane, n-dodecane, isobutene, and cyclohexane. The polymers are modeled as monodisperse chains with 100 repeat units. The average δ values of PE, i-PP, a-PP, PIB, and PI at 300 K are determined as 18.6, 14.9, 14.6, 14.3, and 16.4 MPa1/2, respectively, in a good agreement with experimental data. The δ values of these polymers at various temperatures are also determined. The temperature dependence of δ is fitted to two linear equations, one above and the other below the polymer’s glass transition temperature Tg. The δ values are more sensitive to temperature at T ≥ Tg. The Tg values of the polymers, determined based upon their specific volumes and δ values agree with the experiment qualitatively. The determination of the temperature dependence of δ has a great potential for industrial applications, such as determining miscibility, developing polymeric organogelators as flocculants and oil spill treating agents, and identifying potential solvents and ideal processing temperatures. Methods: The MD simulations are performed using the GROMACS 2022.3 package with optimized potential for liquid simulations-all atom (OPLS-AA) force field parameters. All polymers are built as extended chains using CHARMM-GUI Polymer Builder. Graphical Abstract: (Figure presented.)