Graphene is increasingly being studied for fuel cell catalysis because of its high conductivity, large surface, and chemical stability.[1,2] Graphene oxide (GO) is a monolayer carbon material with carboxyl, hydroxyl and epoxy oxygen functional groups on the surface and edges. These make GO hydrophilic, and the sp2/sp3-hybridization is believed to make it insulating to electrons.[3] Fuel cell membranes have several important requirements. They must be insulating to electrons, have significant ionomer conductivity, provide sufficient barrier to prevent hydrogen/methanol crossover, and have the mechanical strength to be able to support the cell. GO could potentially fulfil these requirements. [4,5] Here, we investigate the suitability of GO for fuel cell membranes by measuring the gas barrier properties, proton conductivity, anion conductivity, electronic conductivity, tensile strength, and fuel cell performance. Initially we reported the characterization of a graphene oxide membrane fuel cell (GOMFC).[6] Free-standing flexible GO membranes were prepared from GO dispersion in water by vacuum-filtration. GO was found to have higher tensile strength and water uptake compared with Nafion. The power density of a fuel cell with a 30 μm thick GO membrane was 35 mW/cm2 at 30°C, despite the fact that the proton conductivity is several orders of magnitude lower than Nafion. The device had a very high open circuit voltage >1 V indicating low crossover. This indicates that thinner membranes compared with Nafion can be utilized, reducing the overall cell resistance. Indeed, by changing the fabrication method to further reduce the thickness of the GO membranes to several microns, we recently achieved improved power density of up to 80 mW/cm2. We also have performed a detailed study on the through-plane conductivity and permittivity of GO membranes over a wide temperature and humidity range.[7] It was found that the proton conductivity is strongly dependant on humidity, and that under dry warm conditions significant electronic conductivity is observed. This opens the opportunity for tuneable mixed ionic-electronic conductivity and may be extremely useful in sensing application. In addition, we performed in-situ scanning electron microscopy (SEM) and electron energy loss spectroscopy (EELS) on humidified and dried GO membranes, directly observing expansion and contraction (“breathing”) of the membranes.[8] Even after drying and under high vacuum conditions, the EELS signal for crystalline ice was observed after freezing, suggesting that water is extremely strongly bound to GO, which may have implications for proton conductivity. Finally, a novel class of alkaline anion exchange membrane (AAEM) is presented, in the form of KOH-modified multilayer graphene oxide paper (GOKOH).[9] SEM investigations showed that the morphology of GO changes after KOH-treatment, whilst X-ray photoelectron spectroscopy (XPS) measurements and X-ray diffraction (XRD) analysis confirmed successful chemical modification. The hydrogen gas permeability was several orders of magnitude lower than conventional polymer-based ionomer membranes. The maximum anion conductivity was 6.1 mS/cm at 70 °C, and the dominant charge carrier was confirmed to be OH− by utilization of anion and proton-conducting blocking layers. The ion exchange capacity was 6.1 mmol/g, measured by titration. A water-mediated reverse Grotthuss-like mechanism is proposed as the main diffusion mode of OH− ions. A prototype AAEM fuel cell was fabricated using a GOKOH membrane, confirming the applicability to real systems.
Read full abstract