AmericanChemicalSociety.com

We have prepared a series of new Nafion-based composite membranes via an in-situ sol-gel reaction of 3-(trihydroxylsilyl)propane-1-sulfonic acid (THSPA) and solution casting method. The hydrophilic nature of THSPSA ensures the compatibility between the inorganic phases and the ionic domains of Nafion matrix, and hence improves the homogeneity of composite membranes. In addition, the resulting sulfopropylated polysilsesquioxane (SiOPS) allows for continuity of proton conduction between the fillers and Nafion through the covalently attached alkyl sulfonic acids.

The formation of polysilsesquioxane in the composites was confirmed by the FTIR spectra. TGA curves indicate that the composite and Nafion membranes have similar onset temperatures around 315 to 325 ^oC corresponding to the degradation of sulfonic acid groups in the side chains. Morphological structures in the composite membranes were examined by TEM, where the phase-separated morphologies with compositional heterogeneity on the scale of 5 - 10 nm can be clearly seen in the composites, similar to those in unmodified Nafion. This is consistent with the speculation that the morphology of Nafion acts as a template for the sol-gel reaction of THSPSA in the ionic domains of Nafion matrix. As the THSPSA content increases, hydrophilic agglomerates with diameters ranging from 20 – 40 nm emerge in the TEM images, which can be ascribed to the formation of large SiOPS nanostructures that grow out of the polar domain of Nafion and become entangled throughout the matrix. These SiOPS aggregates are distributed relatively uniformly in the composite membranes, as verified in the lower magnification TEM images.

The ion exchange capacity (IEC) of the membrane was found to increase rapidly from 0.92 mmol/g of NNS-0 (Nafion) to 1.25 mmol/g of NNS-12 (12 wt.% THSPA). A further increase of THSPSA content from 12 to 20 wt.% does not lead to an obvious rise in IEC. It should be noted that the water uptake of the composite membranes follows the same trend. A gradual increase in water uptake from 19.4 to 28.7% is observed with increasing the THSPSA concentration up to 12 wt.%, and the water uptake levels off thereafter. The saturation of IEC and water uptake can be rationalized in terms of the large SiOPS nano-clusters formed at high precursor concentrations, which trap the inside sulfonic acid groups and make them not easily accessible for interactions with water and Na⁺/H⁺ exchange. To probe the states of water inside the hydrated membranes, ΔH_f for water has been measured using low temperature DSC. In comparison to pure Nafion recasting membrane (NSS-0), it is evident that the composite membranes, especially in the cases of NSS-4 (4 wt.% THSPSA) and NSS-8 (8 wt.% THSPSA), contains more weak bound and free water, which could be correlated to higher concentration of sulfonated groups that interact with water molecules through hydrogen bonding.

Compared with pristine Nafion, the composites exhibit improved proton conductivity, resulting from the increased IEC and water uptake values. The conductivity increases progressively with the THSPSA content and become steady at ~ 0.101 S/cm for NSS-16 and NSS-20, which is in accordance with the IEC and water uptake trends. For purpose of comparison, we have also prepared Nafion/silica composites using sol-gel reaction of tetraethoxylsilane (TEOS). The proton conductivities measured at 30 ^oC under a fully hydrated condition are 0.071 S/cm for 5 wt.% TEOS membrane and 0.07 S/cm for 15 wt.% TEOS membrane. These values are in close agreement with the conductivities previously reported in literature and lower than that of unfilled Nafion having a conductivity of 0.084 S/cm.

The dependence of the proton conductivity of the membranes on humidity and temperature has been investigated. Similar to Nafion, the proton conductivity of the composites increases with temperature and relative humidity. This temperature dependence is associated with the thermal activation process of proton conduction, and the influence of humidity on proton transport is presumably due to the water uptake of the membrane and change of ionic domain sizes at different RHs. Furthermore, the NSS composites were found to exhibit higher conductivity than Nafion over the whole range of humidity and temperature measured at 30 – 90% RH and 30 – 80 ^oC, respectively. The composite membranes show lower activation energy (E_a) values than bare Nafion, suggesting that the proton conduction is more facile in the composites. The decrease of E_a with increasing the THSPSA content and IEC value is noteworthy. Undoubtedly, the presence of sulfonic acid side chains in the polysilsequioxane facilitates transport of protons in the membranes. These results are in sharp contrast to those reported for the conventional Nafion/silica composites, where the fillers were found to disrupt the proton diffusion paths and decrease the conductivities. The observed low E_a values at around 7 kJ/mol imply that conduction at room temperature may occur predominately via the Grotthuss (hopping) mechanism in the NSS composites, which can be idealized as the proton being passed along a chain of rapidly re-orienting hydrogen-bonded. More importantly, the enhancement in proton conductivity becomes clearly pronounced at high temperatures and low humidity. A conductivity of 0.007 S/cm was determined at 80 ^oC and 30% RH for NSS-16 and NSS-20 composites, which is more than two times greater than that of Nafion. At 120 ^oC and 30% RH, NSS-16 and NSS-20 membranes yield conductivities of ~ 0.0102 S/cm, while the conductivity of unfilled Nafion was measured to be around 0.004 S/cm.

This work demonstrates the promising potential of the properly modified Nafion composite membranes for applications in elevated-temperature PEMFCs. A manuscript with the acknowledgement of ACS-PRF support has been submitted to ACS Applied Materials & Interfaces. Matthew R. Gadinski, an undergraduate student who worked on this project for his senior thesis, has been listed as a co-author of this manuscript. A graduate student, Kui Xu, presented his results from this project on 238^th ACS National Meeting in Washington DC on Aug. 16, 2009. The PI co-organized a symposium entitled “Functional Polymer Nanocomposites for Energy Storage and Conversion” in 238^th ACS National Meeting. The objective of this symposium is fully in line with the goal of PRF grant support.