Atomistic and Coarse-Grained Molecular Dynamics Simulation of a Cross-Linked Sulfonated Poly(1,3-cyclohexadiene)-Based Proton Exchange Membrane

Abstract

Atomistic and coarse-grained (CG) molecular dynamics (MD) simulations were conducted for a cross-linked and sulfonated poly­(1,3-cyclohexadiene) (xsPCHD) hydrated membrane with λ­(H2O/HSO3) = 10 and 20. From the atomistic level simulation results, nonbonded pair correlation functions (PCFs) of water–water, water–H3O+ ion, H3O+–H3O+, polymer–water, and polymer–H3O+ ion pairs were obtained and studied. The water self-diffusivity and H3O+ vehicular self-diffusivity were also obtained. Membrane structure was further studied at CG level using a multiscale modeling procedure. Nonbonded PCFs of polymer–polymer pairs were obtained from atomistic simulation of hydrated membrane with λ = 10 and 20. Two sets of CG nonbonded potentials were then parametrized to the PCFs using the iterative Boltzmann inversion (IBI) method. The CGMD simulations of xsPCHD chains using potentials from above method satisfactorily reproduced the polymer–polymer PCFs from atomistic MD simulation of hydrated membrane system at each hydration level. The transferability of above two set of CG potentials was further tested through CGMD simulation of hydrated membrane at an intermediate hydration level (λ = 15). Limited transferability was observed, presumably due to the use of an implicit solvent. Using an analytical theory, proton conductivity was calculated and compared with that from experimental measurement under similar conditions. Good agreement was obtained using inputs from both atomistic and CG simulation. This study provides a molecular level understanding of relationship between membrane structure and water and H3O+ ion transport in the xsPCHD membrane.

Hydrogen adsorption and diffusion in amorphous, metal-decorated nanoporous silica

“…The densities change, as the name implies, for a given eOTiCl 3 content, although there are examples where the density of a "low density" material with 100% eOTiCl 3 end groups is higher than a "medium density" material with 0% eOTiCl 3 end groups due to the high molecular weight of Ti. Accessible volume and surface area for each structure were calculated using geometrical methods that have been used for MOFs [27] and polymer membranes [28,29] in our previous studies. These geometrical methods are based on published approaches [30].…”

Hydrocarbon-based fuel cell membranes: Sulfonated crosslinked poly(1,3-cyclohexadiene) membranes for high temperature polymer electrolyte fuel cells

“…For the latter, hydrogen bonding favored by 2-hydroxycyclohexanesulfonic acid could lead to morphologies conducive to better proton conduction. The effect of microstructures generated by various sulfonating condition has been studied by molecular dynamics simulation [26,28,29]. Good agreement between calculated conductivities and measured values [29] show that the relationship between membrane structure and relevant transport properties can be distilled, at least to a first order, to three key factors: acidity, confinement, and connectivity [30].…”

Evaluation of the Moisture Effect on the Material Interface Using Multiscale Modeling

“…MD simulations have been applied to study the molecular behavior of different substances as a powerful numerical modeling technique [46][47][48][49][50][51][52][53][54][55][56][57][58][59][60][61]. According to the previous experimental studies, the critical region of the FRP debonding from substrate under moisture attack locates at the interface between epoxy and substrate.…”