Alternative Membranes for Low-Temperature Fuel Cells

Proton transfer in aqueous environments

Water in Nafion continues to be the preferred media for proton transfer in hydrogen fuel cells. There are several reasons for this: 1) water is non-toxic, 2) offers relatively good conductivities, 3) self-regenerating, 4) non-poissonous towards the catalyst. A major drawback however, is its relatively low boiling point, which prevents the use of water in open fuel cells above 100oC. Higher temperatures are necessary to achieve greater conductivities. It is important to understand proton conductivity in aqueous environments so that new media can be designed that can operate at higher temperatures.

Recent advances in computer simulations are beginning to address the multi-scale nature of the phenomena, including algorithms for chemical bond cleavage and formation, long integration times of the equations of motion, and the large numbers of atoms needed to obtain converged diffusion coefficients over sufficiently large volumes. These new algorithms are allowing a more fundamental understanding of the proton transfer process in aqueous solutions.

Theory indicates that quantum effects are dominant in aqueous proton transfer events owing to the small mass of the proton, the low energy barriers involved in proton transfer in water, and the relatively large zero point energies. The overall process can be subdivided into consecutive steps of dissociation, reaction, complex reforming steps, and diffusion. In the early steps, a hydronium species (H3O+) transfers a proton via one of its hydrogen bonds to a first solvation water molecule (Figure 1). Any of the protons in the newly formed charged complex may in turn be transferred to a neighboring molecule, with the excess proton constantly changing identity, and migrating throughout the hydrogen bond network at a rate considerably greater than conventional diffusion, the Grotthuss mechanism.

Figure 1: Proton diffusion pathway in water estimated using Quantum Hopping Molecular Dynamics at 300 K. Yellow indicates events for which the transfer occurred without a barrier (purely quantum event). Magenta depicts events for which transition state theory applies.The vehicular diffusion of the H3O species, diffusion alone, is depicted in cyan. Notice how short vehicular diffusion is compared to proton transfer events, either classical transition state events or quantum tunneling events.

Proton transfer in non-aqueous solvents and anhydrous environments

Several types of "water free" proton conducting membranes that incorporate quaternary nitrogen atoms are under investigation: a) polymeric materials with heterocyclic pending groups, such as imidazole, pyrazole or benzimidazole; b) new dendrimer-PTFE copolymers combining hydrophilic dendrimers with hydrophobic linear polymers; and c) novel Bronsted acid based ionic liquids, made by combining organic amines above their melting temperatures with bis (trifluoromethanesulfonyl) amide (HTFSI,these are electroactive for H2 oxidation and O2 reduction at a Pt electrode under non-humidifying conditions at moderate temperatures (ca. 130oC) (Figure 2), d) most promising are simple protic ionic liquid mixtures such as a 4:6 mixture of methyl and dimethyl ammonium nitrate that become superionic at 25oC. At this temperature conductivities are similar to Nafion at 80oC, 150 versus 100 mS/cm. At 100oC, the conductivity is 470 mS/cm, a factor of 4 larger than the conductivity of Nafion.

Figure 2: Examples of proton acceptor for a Bronsted acid HTFSI for non-aqueous proton transfer membrane media.

To explain this superionic proton conductivity we have studied the quantum mechanical barrier for proton transfer in the methyl-ethyl ammonium nitrate system (Figure 3). For distance lower than 2.5 Angstroms proton transfer is barrier less that may help explain the superionic nature of this ionic liquid.

Figure 3: Proton transfer barriers as a function of donor-acceptor (N...O) distance (in Angstroms) in the methylethylammonium nitrate ionic liquid.

Other alternative amine systems under study are:
  • Organic bases (such as cyclic amines and heterocyclic compounds with multiple substituents, including but not limited to F, O, S, N, P)
  • Polymeric versions of the most promising organic bases with organic acids (polyphosphoric acid, sulfonic acid, etc.)
  • Inorganic systems containing high density of hydrogen donor/acceptors but low-lying temperature/pressure phase transitions.

    Personnel: Dr. Mario Blanco