Empirical potential structure refinement of semi-crystalline polymer systems: polytetrafluoroethylene and polychlorotrifluoroethylene.

Empirical potential structure refinement (EPSR) simulations are performed on total neutron scattering data from powder samples of polytetrafluoroethylene (PTFE) and polychlorotrifluoroethylene (PCTFE), both at 300 K. Starting from single strands of polymer consisting of between 30 and 60 monomers of tetrafluoroethylene and chlorotrifluoroethylene in each case, hexagonal simulation cells are constructed consisting of an array 25 (5×5) such strands placed on a hexagonal lattice. Allowed simulation moves are polymer translation moves along all three Cartesian axes, whole polymer rotations about the polymer axis, and individual atom moves within each polymer. For PTFE a number of Bragg peaks are visible in the scattering data and these are found to be consistent with a lattice spacing a(=b) = 5.69(1) Å with a dihedral angle along the (helical) chain of 166° which gives a repeat distance along the chain (c-axis) of ~19.6 Å. The positions of the Bragg peaks are well reproduced by this model, although there is a mismatch in the amplitudes of some of the higher order reflections between simulation and data. For PCTFE there is only one visible Bragg peak (100) which is well reproduced by a hexagonal lattice of atactic parallel polymers with a spacing of a(=b) = 6.37(1) Å. In this case the absence of distinct reflections along the polymer c-axis makes characterization of the internal dihedral angle difficult, but a model with a dihedral angle of 166° was less successful at fitting the diffuse scattering than a model where this angle was set to 180°, giving a nearly straight trans (zig-zag) structure. For PCTFE little change in structure could be discerned when the material was heated to 550 K, apart from a slight increase in lattice spacing. In both cases there is substantial diffuse scattering between the Bragg peaks, and this is correctly replicated by the EPSR simulations.