Development of cost effective materials for use as hydrogen purification membranes

Abstract

The demand for hydrogen is currently over 50 million tons/year and ever increasing due to its role as a critical reagent in petrochemical refining with a possibly larger future as an energy carrier in a hydrogen economy. Greater than 95% of H2 is produced through steam reforming of hydrocarbons, and membrane technologies offer the potential to greatly reduce the cost of H2 purification over conventional separation strategies. The goal of this thesis was to further develop a class of novel, earth abundant membranes for use in hydrogen purification and production. The first investigation sought to establish the full potential of the group-V body centered cubic (BCC) metals for use as hydrogen permeable membranes. V, Nb, and Ta all have significantly higher hydrogen permeability than Pd, but negligible ability to dissociate H2 into atomic hydrogen, and therefore require the application of a H2 dissociation catalyst. 100 nm Pd films were sputtered onto both sides of BCC metal foils to form composite membranes. The measured H2 permeability of all three metals exceeded previously reported values and closely approach theoretical limits. However, the stability of each membrane varied significantly due to unique failure mechanisms. The Pd/V membranes were mechanically robust but failed rapidly due to a combination of Pd-V interdiffusion and high susceptibility to oxidation. The Pd/Ta membranes were the most resilient to oxygen, but their mechanical integrity was relatively poor and they failed within 48 hours due to Pd-Ta interdiffusion. In contrast, Pd/Nb membranes exhibited high permeability throughout the 168 hours of testing, with no Pd-Nb interdiffusion observed. The decline in permeability observed during testing was attributed to partial Pd delamination as a result of membrane deformation. These results provide pathways for further development of these membranes. For example, preliminary experiments demonstrated that Mo2C could serve as an effective interdiffusion barrier, extending the lifetime of Pd/V membranes from < 24 hours to > 100 hours. The second investigation was aimed at developing cost effective and earth abundant H2 dissociation catalysts to replace or reduce the use of platinum group metals (PGMs). The electrochemical analysis of the hydrogen evolution reaction (HER) was used as a surrogate to screen potential candidates as catalysts for the gas-phase H2 dissociation reaction. The order of electro-catalytic ability was first established as Pt > Pd > Mo2C > Mo > V > FTO. The co-catalytic abilities of PGMs were then investigated by applying monolayer levels of Pt and Pd via physical vapor deposition. It was found that the application of just 1 nm of Pt or 2.5 nm of Pd enabled materials like Mo2C and FTO to display HER onset potentials equivalent to bulk Pt or Pd, respectively. Preliminary evidence of the efficacy of this approach was demonstrated when Mo2C/V membranes modified by the application of a < 3 nm thick Pt co-catalyst showed enhanced permeability over uncoated membranes at temperatures below 600 °C. Finally, long term cycling of working electrodes in sulfuric acid against a Pt counter electrode showed evidence of Pt migration and deposition onto the working electrode. After several hundred cycles the HER onset potential of Mo2C evolved to that of bulk Pt, whereas no change in HER kinetics was observed when electrodes were cycled against a carbon counter electrode. Moreover, the improved level of activity was retained when the electrode was removed and tested in fresh solution against a carbon counter electrode. It is suggested that this approach may be a potentially facile and cost effective method for applying minute levels of Pt with submonolayer control, with in situ monitoring of HER onset potential providing a direct measure of the amount of Pt deposited.