On the basis of our solid-state NMR characterization of dynamics in two model salts, we draw the analogy to the fuel cell membrane candidate, phosphoric acid-doped poly(benzimidazole), and conclude that phosphate anion dynamics contribute to long-range proton transport, whereas the mobility of the polymer itself is not a contributing factor. This is contrasted with emerging membrane candidates, which rely on fully covalently bonded acid donors and acceptors, and target high-temperature PEM fuel cell operation in the absence of liquid electrolyte. The hydrogen-bonding structures of benzimidazolium phosphate and benzimidazolium methane phosphonate are established using X-ray diffraction paired with solid-state 1H DQF NMR. By comparing the dynamics of the phosphate and methane phosphonate anions with the dynamics of imidazolium and benzimidazolium cations, the relative importance of these processes in proton transport is determined. The imidazolium cation is known to undergo two-site ring reorientation on the millisecond time scale. In contrast, it is shown here that the benzimidazolium rings are immobile in analogous salts, on a time scale extending into the tens of seconds. Therefore, we look to the phosphate anions and demonstrate that the time scale of tetrahedral reorientation is comparably fast (50 ms). Moreover, the 31P CODEX NMR data clearly indicate a four-site jump process. In contrast, the methane phosphonate undergoes a three-site jump on a slower time scale (75 ms). A mechanism for a zigzag pathway of proton transport through the phosphonate salt crystallites is developed based on the 31P CODEX and 1H variable-temperature MAS NMR data.

Hydrogen bonding plays a critical role in proton-conducting polymers, as it provides the network necessary for structural (Grotthus mechanism) diffusion. This network must be both pervasive and dynamic in order for long-range proton transport to be achieved. The structural motifs must be understood, even in amorphous materials, and moreover, the lattice energies in the structure must be low enough to allow rearrangement and mobility. To this end, a novel proton-conducting candidate, 1,10-(1-H-imidazol-5-yl)decanephosphonic acid and its HBr doped counterpart are considered from the molecular level as potential proton-conducting membranes. The use of high-resolution solid-state 1H NMR to elucidate structure and dynamics of such systems is highlighted in this material. We compare our molecular-level results to macroscopic probes of proton transport in related polymers, achieved using impedance spectroscopy.

No abstract

Proton dynamics in polymer electrolyte membranes are multifaceted processes, and the relative contributions of various mechanisms can be difficult to distinguish. Judicious choices of model systems can aid in understanding the critical steps. In this study, we characterize anion dynamics in a series of benzimidazole-alkyl phosphonate salts, and compare those dynamics to a membrane prototype, built on a decane backbone. The series of salts are characterized, using high resolution (1)H solid-state magic angle spinning (MAS) NMR, DQ MAS NMR, and (31)P centreband-only detection of exchange (CODEX) NMR spectroscopy, to determine the influence of the nature of the alkyl group on the rates and geometries of anion dynamics, and overall proton exchange processes. The alkyl group is shown to slow the correlation times for anion reorientation, when compared at ambient temperature. However, it is also apparent that the lowered lattice energy of the salt lowers the activation energy and allows good dynamics at intermediate temperatures in both the benzimidazolium ethylphosphonate and in the HBr adduct of 1,10-(1-H-imidazol-5-yl)decanephosphonic acid (Imi-d-Pa).

Imidazole phosphate and phosphonate solid acids model the hydrogen-bonding networks and dynamics of the anhydrous electrolyte candidate for proton exchange membrane fuel cells. Solid-state NMR reveals that phosphate and phosphonate anion dynamics dominate the rate of long-range proton transport, and that the presence of a membrane host facilitates proton mobility, as evidenced by a decreased correlation time of the composites (80 ± 15 ms) relative to the pristine salt (101 ± 5 ms). Benzimidazole ethylphosphonate (Bi-ePA) is chosen as a model salt to investigate the membrane system. The hydrogen-bonding structure of Bi-ePA is established using X-ray diffraction coupled with solid-state (1)H-(1)H DQC NMR. The anion dynamics has been determined using solid-state (31)P CODEX NMR. By comparing the dynamics of ethylphosphonate groups in pristine salt and membrane-salt composites, it is clear that the rotation process involves three-site exchange. Through data interpretation, a stretched exponential function is introduced with the stretching exponent, β, ranging 0 < β ≤ 1. The (31)P CODEX data for pristine salt are fitted with single exponential decay where β = 1; however, the data for the membrane-salt composites are fitted with stretched exponential functions, where β has a constant value of 0.5. This β value suggests a non-Gaussian distribution of the dynamic systems in the composite sample, which is introduced by the membrane host.