High-temperature pulsed field gradient nuclear magnetic resonance self-diffusion measurements of n-alkanes in MFl-type zeolites

Abstract

The coefficients of intracrystalline self-diffusion of t h e n-alkanes from propane to n-hexane adsorbed in zeolite ZSM-5 are studied by means of the pulsed field gradient (PFG) NMR method with field gradient amplitudes up to 25 T m-' over a temperature range from -20 to +380"C. The diffusivities are found to decrease monotonically with increasing chain lengths. They are in satisfactory agreement with the results of quasielastic neutron scattering and MD simulations as well as with non-equilibrium measurements by the frequency response and single-step adsorption technique. However, in comparison with zero-length-column (ZLC) desorption and permeation studies, these data, though exhibiting the same trends of chain length dependence, are ca. two orders of magnitude larger. A s expected, the diffusivities are intermediate between those for zeolite NaX and NaCaA, exhibiting, however, the smallest activation energies. This indicates that t h e differences in the potential energies of t h e adsorbate molecules on their intracrystalline trajectories are in general smaller than in zeolites NaCaA and NaX.

Intracrystalline Diffusion in Zeolites Studied by Neutron Scattering Techniques

“…The agreement between QENS [22] and PFG NMR data [23] is reasonable. The same ZSM-5 sample was used for methane [24], but crystals of a different origin were used for longer alkanes.…”

“…The ZLC (zero-length column) diffusivities [28] are systematically 2 or 3 orders of magnitude lower than the QENS values. Figure 11 Diffusion coefficients of n-alkanes obtained at 300 K by various experimental and theoretical methods in MFI-type zeolites: ( ) MD simulations [25], ( ) hierarchical approach [26], (+) QENS [22], ( ) PFG NMR [23], ( ) single-crystal permeation [27], and ( ) zero-length column (ZLC) [28].…”

Positron Emission Profiling Study of the Diffusion of 3-Methylpentane in Silicalite-1

“…The activation energies of diffusion determined from the slope of the Arrhenius plot at high and zero loadings are equal to 46 AE 6 kJ=mol and 44 AE 4 kJ=mol respectively, meaning that for 3-methylpentane in silicalite, the activation energy does not depend on the pore occupancy. Heink et al [30], using PFG NMR, observed a similar picture for C 3 -C 6 n-alkanes in silicalite. For these alkanes, the selfdiffusivities decrease with the loading, while the activation energies were found to be independent.…”

Using temporal analysis of products and flux response technology to determine diffusion coefficients in catalytic monoliths

“…These changes may be associated with the macroscopic techniques observing other effects such as desorption during measurements as well as the actual diffusion of the molecules in the micropores. These additional effects have been shown to strongly influence the diffusivities observed (Crank 1979;Ruthven, 1988b andVan den Begin et al, 1989;Heink et al, 1992;Jobic et al, 1992;Hufton and Danner, 1993;Silva and Rodrigues, 1996;Sun et al, 1996;Cavalcante et al, 1997;Nijhuis et al, 1999;Loos 2000;Ruthven and Brandani 2000;Gobin 2006;Gunadi and Brandani 2006;Richardson et al, 2008;Karge et al, 2008;Guimarães et al, 2010).…”

“…Table 4 gives the activation energies for diffusion determined by other techniques to compare with this study. A large variation between the methods is observed dependant on whether the technique is microscopic or macroscopic in nature (Eic and Ruthven, 1988b;Heink et al, 1992;Sun et al, 1996;Karge et al, 2008). The microscopic techniques generally result in high diffusivities and low activation energies for diffusion.…”