No abstract

Graft copolymers with ionic backbones and hydrophobic fluorinated side chains have been prepared by using lithiated poly(2,6-dimethyl-1,4-phenylene oxide) (PPO) as a macroinitiator for anionic polymerization of 4-fluorostyrene. After grafting of the poly(4-fluorostyrene) (PFS) side chains, the PPO backbone was selectively sulfonated using trimethylsilylchlorosulfonate under mild and controlled conditions. Microscopy of solvent cast membranes revealed copolymer self-assembly into remarkably regular and well-ordered morphologies which, depending on the molecular structure, included lamellar and cylindrical arrangements of the proton conducting ionic nanophases. Thermal analysis indicated separate glass transitions of the PFS and PPO phases, and high thermal degradation temperatures of the membranes at approximately 220 and 300 °C for the H+ and the Na+ forms, respectively. The proton conductivity of fully hydrated acidic membranes was similar to that of Nafion, reaching above 0.2 S cm−1 at 120 °C. Compared at the same ion exchange capacity, the proton conductivity of the graft copolymer membranes was two times higher than that of a membrane based on an ungrafted sulfonated PPO. The study showed that it is possible to tailor and prepare proton-exchange membranes with well-ordered morphologies and high proton conductivity by employing graft copolymers with a sulfonated backbone bearing hydrophobic side chains.

In the present work we have performed classical molecular dynamics modelling to investigate the effects of different types of forcefields on the stress-strain and yielding behaviours in semi-crystalline lamellar stacked linear polyethylene. To this end, specifically the all-atomic optimized potential for liquid simulations (OPLS-AA) and the coarse-grained united-atom (UA) force-fields are used to simulate the yielding and tensile behaviour for the lamellar separation mode. Despite that the considered samples and their topologies are identical for both approaches, the results show that they predict widely different stress-strain and yielding behaviours. For all UA simulations we obtain oscillating stress-strain curves accompanied by repetitive chain transport to the amorphous region, along with substantial chain slip and crystal reorientation. For the OPLS-AA modelling primarily cavitation formation is observed, with small amounts of chain slip to reorient the crystal such that the chains align in the tensile direction. This force-field dependence is rooted in the lack of explicit H-H and C-H repulsion in the UA approach, which gives rise to underestimated ideal critical resolved shear stress. The computed critical resolved shear stress for the OPLS-AA approach is in good agreement with density functional theory calculations and the yielding mechanisms resemble those of the lamellar separation mode. The disparate energy and shear stress barriers for chain slip of the different models can be interpreted as differently predicted intrinsic activation rates for the mechanism, which ultimately are responsible for the observed diverse responses of the two modelling approaches.

We study the phase stability and martensitic transformation of orthorhombic and monoclinic polyethylene by means of density functional theory using the nonempirical consistent-exchange vdW-DF-cx functional [Phys. Rev. B 89, 035412 (2014)]. The results show that the orthorhombic phase is the most stable of the two. Owing to the occurrence of soft librational phonon modes, the monoclinic phase is predicted not to be stable at zero pressure and temperature, but becomes stable when subjected to compressive transverse deformations that pin the chains and prevent them from wiggling freely. This theoretical characterization, or prediction, is consistent with the fact that the monoclinic phase is only observed experimentally when the material is subjected to mechanical loading. Also, the estimated threshold energy for the combination of lattice deformation associated with the T1 and T2 transformation paths (between the orthorhombic and monoclinic phases) and chain shuffling is found to be sufficiently low for thermally activated back transformations to occur. Thus, our prediction is that the crystalline part can transform back from the monoclinic to the orthorhombic phase upon unloading and/or annealing, which is consistent with experimental observations. Finally, we observe how a combination of such phase transformations can lead to a fold-plane reorientation from {110} to {100} type in a single orthorhombic crystal.

No abstract