Morphology Evolution and Adsorption Behavior of Ionomers from Solution to Pt/C Substrates

Abstract

Coarse-grained molecular dynamics simulations were performed to understand the morphological evolution and adsorption mechanism of Nafion ionomers from the aqueous solutions to the Pt/C substrate surface under various solution compositions and substrate properties. We found that the ionomer coverage did not increase with the increasing ionomer-to-carbon ratio but was related to the size and concentration of the ionomer aggregates, following the Langmuir adsorption model that shows a wettability switching behavior due to their changed morphology from solution to the surface. Ionomer aggregates in the solution tended to unfold and spread on the carbon substrate rather than Pt particles, although the cylindrical ionomer aggregates were easily attracted by Pt particles initially due to their hydrophilic ionic shells. The smaller Pt particles had a greater effect on ionomer adsorption. With the increasing number of Pt particles, ionomer coverage increased first and then decreased, depending on whether there was enough carbon surface to anchor the ionomer backbone. A balanced Pt/C ratio and the appropriate distribution of the Pt particles were required for tuning the ionomer coverage and distribution toward the design of the catalyst ink structure to improve the power performance.

Interfacial Distribution of Nafion Ionomer Thin Films on Nitrogen-Modified Carbon Surfaces

“…On the as-deposited carbon surface, the main chain of the Nafion ionomer adsorbs onto the hydrophobic carbon surface via hydrophobic interactions, ,, leading to the formation of a thin ionomer adsorption layer. In the case of the N-modified carbon surface, the SO 3 – groups in the side chains of the Nafion ionomers are bound to the positively charged surface, which is created by the N functional groups, via Coulombic interactions.…”

Deep Learning to Reveal the Distribution and Diffusion of Water Molecules in Fuel Cell Catalyst Layers

“…As shown in Figure , the catalyst point cloud could be expressed as { N 1 * P 1 }, where N 1 represented the total number of points representing Nafion and Pt/C atoms in a case, and P 1 mainly included the spatial coordinates of every Nafion, Pt and Carbon atoms, water contents, and the parameters for the interaction strength of every Nafion and Pt and Carbon atoms with water, , which could be obtained by the MD simulation as shown in Table S2. LAMMPS software could directly mesh the MD simulation results and output high-resolution points by adding the values over the atoms in each mesh at each time step, then averaging the per-mesh values over long timescales, corresponding to the centroids of each mesh.…”

“…A series of calculations were performed to obtain a wide range of contact angles that Martini bead types represent so that the wettability of the substrate is quantified. The setting details of the interaction parameters refer to our previous study. , The carbon support was represented by a hexagonal net layer with a surface area of 200 × 200 Å 2 for all of the models. The relation between the surface area and the mass of carbon was taken from the experimental data of Vulcan XC-72 ≈ 220 m 2 /g, and therefore the mass of the carbon substrate in the present study is 110,000 g/mol.…”

Prediction of water transport properties on an anisotropic wetting surface via deep learning

“…We used the molecular dynamics method to construct a liquid flow system, where Pt/graphene acted as substrates and their anisotropic wetting surfaces can significantly affect the interfacial transport properties of liquids. [28][29][30][31][32][33] The developed modified Martini force field [34][35][36][37] was used to construct water and Pt/ graphene surfaces, which are shown in Fig. 1.…”