Abstract
Commercial XC-72 carbon black, multi-wall carbon nanotube (MWCNT), and bamboo charcoal were used as carbon supports for Pt catalyst for application in proton exchange membrane fuel cell (PEMFC). Due to the unique morphology of carbon nanotube, there are less geometric barriers than XC-72 for deposition of Pt. Aggregation of platinum particles may be decreased by using carbon nanotube as the support. Surface treatment on MWCNTs has been studied for many years. In order to get more active sites on the surface of MWCNTs, a sonochemical treatment with sulfuric acid and nitric acid was made. By this process, functional groups such as carbonyl (-CO), hydroxylic (-COH), and carboxylic (-COOH) can be created on the surface of carbon nanotubes. Bamboo charcoal has a very large surface area, and is considered as a good candidate as the support. There are many pores on the surface to make easier gas diffusion. Chloroplatinic acid was used as the precursor, and ethylene glycol was used as the solvent and reducing agent to get well-dispersed platinum nanoparticles on the carbon supports. The particle size and dispersion of platinum particles are two key factors to determine the performance of catalysts. In this experiment, platinum particle size was controlled by changing the loading percentage of platinum. The deposition of Pt was characterized by X-ray diffraction, thermal gravimetric analysis, and transmission electron microscopy. It was found that platinum particles were well-dispersed on all carbon supports. The average diameter of platinum particles deposited on XC-72 with 5 wt % Pt loading was lower than 3 nm. The performance of single-cell PEMFC using the Pt deposited on various supports was evaluated, as follows: When using XC-72 as the support and controlling the Pt loading to 5 wt%, the cell power was close to 0.17 watt, higher than that of the cell using Pt/XC-72 as catalyst with 10 wt% or 20 wt% Pt loading, 0.16 watt. The reason was that Pt particles could be well dispersed on XC-72 and the size of Pt particles was lower than 3 nm when the Pt loading was controlled to 5 wt%. The cell power was also higher than that using commercial E-TEK as catalyst, 0.15 watt, due to the same reason. When using M-CNT as the support and controlling the Pt loading to 5, 10, or 20 wt%, the power of the cell was close to 0.12 watt, because the dispersion of Pt particles could not be improved much, and because of residual impurities in the CNT sample and smaller surface area of Pt on the carbon materials. When using bamboo charcoal as the support and controlling the Pt loading to 20 wt%, the cell power was close to 0.12 watt, being possibly because of the impurities in the bamboo charcoal and the large size of Pt particles.