NMR cryodiffusometry determination of pore utility differentials for liquid-phase diffusion within alumina, catalyst support pellets

• Cryodiffusometry assessed efficacy of controlled introduction of templated porosity. • Contributions of individual void space elements to overall mass transport determined. • Independent validation of presence of dead volume via alternate measurements. • Computationally-efficient model constructed that predicted mass transport rates. • Origin of dead volume determined as resultant of localised poor connectivity. Intelligent design of optimal void spaces for diverse applications of porous media remains a still outstanding objective. The aim here was to predict the tortuosity for liquid-phase diffusion within complex, bidisperse, porous alumina pellets, and, thereby, assess the degree of control achieved via a porogen-based synthesis. NMR cryodiffusometry was used as a novel means to implement the so-called “sifting strategy” to identify the key aspects of the void space to include in a more efficient, minimalist, pore structural model for materials unsuited to a “brute-force” modelling approach. The cryodiffusometry showed, through the particular mathematical form of the data obtained, that the predominant geometry of the void space was a random cluster, but also detected the presence of some “dead volume” that provided little, or no, apparent contribution to the overall mass transport. This particular feature of the void space was corroborated by applying the concept of bound volume index to mercury porosimetry retraction data. A minimalist model, constructed incorporating these two key aspects of the void space, was found to be fully predictive of mass transport rates of water, or cyclohexane, within the alumina pellets. Seeded percolation analysis of gas sorption scanning curves demonstrated the dead volume was associated with particular pore space regions with low accessibility, which was consistent with previous Monte-Carlo simulations of self-diffusion on random cluster models and TEM data. The theoretical framework involving a random cluster with dead volume provided a more comprehensive account of the diverse experimental findings than a complementary critical path analysis-based approach.