No abstract

No abstract

There is only limited experimental data and theoretical treatment available in the literature on hydrogen sorption and diffusion in liquid metals, in stark contrast with that in solid metals. This paper utilizes our predictive phenomenological model requiring minimal input, the Pauling Bond Valence-Modified Morse Potential (PBV-MMP) model, for estimating the thermodynamic and kinetic parameters of hydrogen solution and diffusion in liquid metals treated as quasi-crystalline. The PBV-MMP model is a refinement of the Unity Bond Index-Quadratic Exponential Potential (UBI-QEP) model for estimating the energetics of solid metal surface reactions. The sequential kinetic steps of hydrogen dissociative surface adsorption on the feed side, subsurface penetration, and atomic interstitial diffusion in the bulk, followed by these steps in reverse on the permeate side, are thus treated via the PBV-MMP model within Eyring’s transition-state theory framework, while the entropic changes are evaluated via Eyring’s free volume model. Our predictions agree with experimental results for different liquid metals reported in the literature as well as with our results for hydrogen sorption and permeation in our recently reported sandwiched liquid metal membrane (SLiMM).

No abstract

SignificancePalladium-based membranes are currently the most advanced membranes for hydrogen separation and are on the verge of practical application. However, the search for alternative membranes continues in an effort to lower their cost and susceptibility to poisons. Here for the first time we report a novel sandwiched liquid metal membrane (SLiMM) for hydrogen separation. Permeation experiments indicate that the Ga/SiC SLiMM has a permeability of 2:75 310 27 mol=ms Á Pa 0:5 at 5008C, which is 35 time higher than that for Pd under similar conditions. This promises a potential for application of SliMM in hydrogen purification.