(551f) Predicting ? for Polymer Blends with “Morphing” Simulations
AIChE Annual Meeting
2020
2020 Virtual AIChE Annual Meeting
Materials Engineering and Sciences Division
Polymer Thermodynamics and Self-Assembly: Predicting Properties
Wednesday, November 18, 2020 - 9:15am to 9:30am
The Flory Huggins interaction parameter χ governs phase behavior in polymer blends and block co-polymers. χ is a measure of the excess free energy required to blend polymers. We have previously used molecular dynamics (MD) simulations and thermodynamic integration during the “morphing†of one species to another, to compute the excess free energy of mixing in a bead-spring model of polymer blends, for chains with different stiffness or Lennard-Jones interactions. In this work, we present two studies which use the morphing method to: (1) explore the effect on χ of chain architecture in bead-spring chains, and (2) determine χ for chemically realistic polymer blends.
Chain architecture affects how chains pack and interact in the melt, which can significantly influence χ. To explore this, we investigate χ for blends with different architectures of flexible bead-spring chains. We examine blends in which both chain species have a polypropylene-like structure, but one species has beads with a weaker interaction: either the side beads (case 1), main chain beads (case 2), or branch point beads (case 3). We find χ for all three cases, for which random mixing models would give identical results. χ is largest for weakened side beads, which are more accessible to close contacts with other beads. These systems provide an appealing test for PRISM predictions of structure and miscibility in polymer blends. We find PRISM predictions for χ are only qualitatively consistent with our morphing results, and troublingly sensitive to the initial guess needed to solve the integral equations.
Using united-atom MD simulations and extensions of the morphing idea, we compute χ for real polymer blends, including (1) polyethylene / polyethylene oxide, (2) polystyrene / poly(2-vinyl pyridine), (3) polyisoprene / saturated polyisoprene and (4) polystyrene / poly(α-methyl styrene). These examples require different kinds of morphing: morphing LJ interactions and partial charges of atoms (cases 1 and 2), transforming double bond to single bond (case 3) and making atoms disappear (case 4). Our χ values from simulations are in reasonable agreement with experiment, but are sensitive to force field parameters used in the simulation.
Chain architecture affects how chains pack and interact in the melt, which can significantly influence χ. To explore this, we investigate χ for blends with different architectures of flexible bead-spring chains. We examine blends in which both chain species have a polypropylene-like structure, but one species has beads with a weaker interaction: either the side beads (case 1), main chain beads (case 2), or branch point beads (case 3). We find χ for all three cases, for which random mixing models would give identical results. χ is largest for weakened side beads, which are more accessible to close contacts with other beads. These systems provide an appealing test for PRISM predictions of structure and miscibility in polymer blends. We find PRISM predictions for χ are only qualitatively consistent with our morphing results, and troublingly sensitive to the initial guess needed to solve the integral equations.
Using united-atom MD simulations and extensions of the morphing idea, we compute χ for real polymer blends, including (1) polyethylene / polyethylene oxide, (2) polystyrene / poly(2-vinyl pyridine), (3) polyisoprene / saturated polyisoprene and (4) polystyrene / poly(α-methyl styrene). These examples require different kinds of morphing: morphing LJ interactions and partial charges of atoms (cases 1 and 2), transforming double bond to single bond (case 3) and making atoms disappear (case 4). Our χ values from simulations are in reasonable agreement with experiment, but are sensitive to force field parameters used in the simulation.